pazza83
commented
3 years ago Following discussion today on Keybase and in the thread Why does Bisq need a new GUI? it would be great to see if we can achieve a rough consensus about how to move forward.
My proposal is as follows:
@dukke to be complete Stage One: Coding and onboarding
This will involve:
- Performance test FXML vs Plain Java (there is a certain amount of coding needed as recommend by @ripcurlx prior to changing the design)
- New user onboarding process, see @pedromvpg's designs here: https://xd.adobe.com/view/a83c2327-4730-4ec2-8938-e318b2749588-fd6f/
Rationale
The rationale for getting started is as follows:
- Bisq is in need of a UI Developer with the skills and availability to contribute to Bisq. @dukke is available and has had discussions with @chimp1984, @pedromvpg, and myself.
- @dukke will be in a much better position to estimate costings for the project once they have completed the coding required for on-boarding.
- The on-boarding can be improved to make it more user friendly.
- Even if Bisq decided not to proceed with the planned UI redesign having a UI Dev on board that could work with us to implement changes to the existing UI would be great.
Risks of not doing anything now
The risks of not doing anything now is that @dukke will take on another project and become unavailable. Bisq would still be in need of a UI developer and it might be a while before one is found.
Vote
I am keen to get back to @dukke next week.
Please can people vote on this comment for the proposal: @dukke to complete Stage One: Coding and onboarding.
👍 - Yes, lets start the UI project on stage one then review
👎 - No, lets not start the UI project
Any other comments welcome 😃
Description
Implement new-user-onboarding and new UI design as discussed in #47
Rationale
Bisq is a P2P exchange that is accessed via a software client that while extremely functional has remained largely unchanged from a user interface and user experience perspective. In order to increase the accessibility of Bisq for new and existing users the user interface and user experience should be redesigned.
Goals
The goals of the redesign should be to:
Why now?
The UI designs, being done by @pedromvpg, have in a state nearing completion for a while, the barrier to implementing them was lack of UI developer resources. @chimp1984 has recently been in touch with a developer that is keen to get involved in a UI resign. Having a dev in place and most of the designs already complete means it is a good time to start a UI redesign.
The risks of not doing it now are that Bisq may struggle to have the UI developer resources at hand to implement the project.
Criteria for delivery
The project is fairly large and will be implemented in the following stages:
A rough project plan can be seen here: https://docs.google.com/spreadsheets/d/1RvjngdWhKX8L47tLkEP9eiJjfSb-SyUWvHnDYZP5YcQ/edit#gid=1115838130
Coding and onboarding stage
The first two aspects will be:
The reason for the above two being first is that they can be designed and implemented first without effecting the other aspects of the client.
Client redesign stage
Once the above have been successfully completed the project will move into redesigning the user interface of the following sections:
The current designs for the above by @Pedromvpg can be seen here: https://xd.adobe.com/view/b01dfd7a-3f79-4744-8df9-08394d2ea1ea-e54e/grid/
The above sections will be more clearly outlined following completion of the intial stage.
It is expected that once the client has been redesigned it will be necessary to resign the home screen to pull in the necessary data to make trading on Bisq an effective, efficient and satisfying experience.
The final stage will be allow for a period of user review and feedback before implementing any changes.
Measures of success
The success of the project will be measured against the goals:
Objective measurements:
It is preferable to measure goals in objective terms. Goals that can be measured objectively include:
Anecdotal measurements
Where goals can not be measured objectively due anecdotal measurements will be used to assess the changes in:
Risks
Changing the way users interact with Bisq will not come without risks.
I will defer to the devs for risk in implementing a new UI.
Other risks to be considered are:
Risks of project being stalled / delivered incomplete. This will be mitigated by breaking the project into sections and ensuring that any work delivered can be picked up by another contributor.
Design is subjective, there is the risk, or reality, that not all users are going to prefer a new design. This risk will be mitigated by trying to involve users from the outset and giving users the opportunity to express their opinions and input on any new designs.
Tasks
Roles
Estimates
The estimates will be broken down into the various stages.
Each section will have confirmed costs from all contributors involved in the various stages.
The costs for the initial stage of Performance test FXML vs Plain Java will be confirmed by @ripcurlx and the new UI dev.
@chimp1984 previously put the cost for the whole project somewhere between $30-60k USD.
Notes
The new UI dev is new to Bitcoin and the DAO model.
@chimp1984 previously made the following offer:
I have offered to pre fund the work of the UI dev so that the new UI can be completed. It is still to be confirmed how this will work exactly but I think in principle myself, the new UI dev and other contributors are happy with this model, although progressing to a DAO model for the long-term would be preferable.