Introducing Stravatar

Introducing Stravatar

I think Twitch is great, and the product market fit is there. Twitch has very clearly demonstrated value in building online communities of people with a shared interest in gaming.

But this idea isn’t novel or new - online communities with a shared interest in gaming have been around since the early internet via forums and discussion groups. It just so happens that live streaming has risen to become the most popular medium for engaging in discussion around gaming.

By watching a game in real time you get to experience the positive feedback loops as if you were playing it yourself. The more invested you are in the community, the more intense the experience. That’s because when you’re watching together, you feel like you’re playing together.

It seems like the only experience that would be more interesting is if a community played the game together rather than just watching it together. But there’s challenges with this.

Most games do not have great tools for building and maintaining online communities - with the exception of really successful MMOs like WoW. This includes community based rewards like bonus XP, cosmetics and customizations.

Most games don't offer viewership options like a spectator mode - or the capacity for viewership is not big enough to satisfy a larger audience. Sometimes it’s as simple as wanting to support your guild during raid time, but there aren’t enough seats left in the group.

Twitch aims to solve these issues by offering platform tools for community building and live streams to keep communities together while a game is being played.

There’s also scalability issues live streaming platforms solve that would be very hard for games to emulate. It’s much easier to scale live video than game state. Game servers broadcast game state to connected clients but have resource limitations on the number of clients that can be connected at a time.

To serve game state to a large viewership in game you would need to offload broadcasting to a secondary delivery method. Most games receive game state through RUDP so it’s likely the secondary delivery method would proxy the original stream because it’s already as optimized as it can get.

Since game state is very sensitive to latency, the secondary delivery method would need to have similar throughput as the original game server, which could become very expensive at scale.

A less expensive method might be to transcode and broadcast the game state in a format compatible with live video infrastructure. That way connected clients could receive a live stream using a custom decoder to get back an original data format the game engine will understand.

This comes at the cost of a delay in game state delivery, but now we’ve come full circle back to live streaming anyway :D.

Being web based, live streaming platforms are also more accessible then game clients. You can watch the video from any device, listen to the commentary of the live streamer if you aren’t directly watching, or use closed captions when you want to watch it quietly.

So if streaming platforms already have excellent community building, delivery and accessibility, what else is left to improve in online gaming communities?

If you look around on Twitch you’ll see streamers almost always use additional software not available on the streaming platform to improve the experience of the viewer. This includes live streaming overlays like donation event animations, virtual stream avatars, VFX filters, streamer led games like Marbles on Stream, chat bots, and live giveaways.

With so many options available, it seems the only thing left to improve in online gaming communities is viewer experience. The fact is, that if these platforms offered everything a streamer needed to create a comparable experience, this other software wouldn’t exist.

Surely as a platform with unlimited resources, building these missing features back into the service shouldn’t be a problem right? I think it probably comes down to two things. Either these features are not profitable enough to pursue or the monopoly of viewership means there’s no reason to innovate and improve the viewing experience.

So if you want to create a product that improves viewer experience you only need to be as good or better than other companies innovating in this space. Many products are competing in the viewer experience space but none have demonstrated a 10x value needed to emerge as the clear winner.

Stravatar has been in the making since 2015 and is a consolidation of technology and lessons learned from my other projects in the same space.

Without giving too much away, Stravatar offers viewers identity customization options through the use of avatars that can be overlaid on top of a stream. Stravatar also offers viewer rewards in the form of cosmetics for their avatar.

But these are really just surface level features of the core product. Stravatar aims to change key viewership and streaming behaviors to make live streaming a more engaging experience. Very excited to unveil the product and the technology used to build it!