Transforming and improving the communication experience in WorkAdventure

After carefully listening to WorkAdventure’s community and their feedback regarding the interactions between users and its limitations within the app, we decided to adopt a new system that enables bubble discussions with unlimited participants, speaker zones, and a neat and ergonomic experience inside meeting rooms.

WorkAdventure’s main goal is to replicate real-life experiences in virtual worlds that can be used for many purposes, such as workplaces, classrooms, recruitment, and social events.

For that matter, improving the user experience is a significant purpose in recreating and properly emulating the in-person experience.

To achieve this objective, new features involving communications and interactions are being developed:

• Discussion bubbles with unlimited participants.
• Unified interface in the meeting rooms.
• Speaker zones.

Boosting the discussion bubbles: no more limits

The discussion bubbles are triggered every time a user approaches another person with their avatar, automatically starting a video conversation between the interlocutors. 

When two people approach each other to engage in a conversation, their avatars can be spotted inside a bubble, and more people can join their discussion just by going inside this area.

The previous system was based on peer-to-peer communication, which was effective but limited. As a matter of fact, no more than four participants could interact within a discussion bubble.

Nevertheless, the new system will enable discussion bubbles with an unlimited number of speakers, making it no longer necessary to move to a meeting room every time you need to speak to more than 4 people, enhancing the virtual conversation experience.

Unified interface in meeting rooms

Jitsi, WorkAdventure’s previous system for meeting rooms, had an interface of its own. Because of this, every time somebody goes inside a Jitsi meeting room, that interface has to be displayed over WorkAdventure’s visual support, thus providing an uneven visual appearance.

Furthermore, Jitsi meeting rooms had a different text chat that was completely unrelated to WorkAdventure’s original chat support.

With Livekit, the new system, there will be no need to merge different interfaces because Livekit is already a low-level software piece; hence, it doesn’t impose an interface of its own. Thanks to this advantage, the experience inside meeting rooms for a virtual conversation will feel much more neat, unified, and ergonomic.

Speaker zones

On numerous occasions, people find themselves in situations where they have to listen to a speaker who’s reaching out to a large number of people. Lectures, classes, webinars, workshops, and conferences are just a few examples of this case.

WorkAdventure decided to fulfill this need by creating the speaker zones: areas where only the speaker can turn on their camera and microphone, whereas the participants can only stay muted.

The speaker zones powered by Livekit could replicate the conference experience thoroughly:

• The speaker is seen and heard by everybody in the speaker zone.
• The attendees can’t speak.
• The attendees can’t be seen.

This new system will be able to replicate the unidirectional lecture experience, which can be very useful in numerous cases to improve this type of virtual conversation.

Comparing Jitsi and Livekit

Both Jitsi and Livekit work as selective forward units (SFU) that host WorkAdventure’s communication processes, especially the discussion bubbles and the meeting rooms

Reviewing Jitsi

Jitsi is a set of open-source projects that allows you to build and deploy a complete video conferencing solution.

As it was stated, Jitsi powers the meeting rooms and imposes its interface; that’s why the appearance of the meeting looks different compared to WorkAdventure’s look, and the meeting chat is completely unrelated to the original chat support of WorkAdventure.

Peer-to-peer communication

It’s nevertheless important to mention that Jitsi didn’t work as a source and engine for interactions in discussion bubbles. On the contrary, when people approach each other with their avatars, they trigger a video discussion that works on a peer-to-peer basis.

The peer-to-peer system establishes connections between users who can speak and write to each other. This connection only passes through the browsers of the different participants without needing to go through a general server.

Although the peer-to-peer connection is efficient, direct, secure, and fast, it’s difficult to establish this type of interaction between more than 4 users due to its limitations.

Migrating to Livekit

Just like Jitsi, Livekit is an SFU. Its goal is to relay real-time video streams between users. Unlike Jitsi, Livekit is a low-level set of APIs. It does not come with any user interface. In the case of WorkAdventure, this is a good thing because we can integrate the Livekit video streams wherever we want in the WorkAdventure page.

Thanks to the implementation of Livekit, conversations in discussion bubbles will now run through a common server, enabling the possibility of interactions with an extremely high number of participants. Technically, we will be able to scale to several thousand users in the same bubble or meeting, and gracefully evolve from a peer-to-peer setup to Livekit automatically.