XaosPrincess Reflects on High Fidelity: What Went Wrong?

(“We now return you to our regularly scheduled programming…”  If you are looking for my blogposts about the Wuhan coronavirus/2019-nCoV/SARS-CoV-2/COVID-19, please click here. Thanks!)


XaosPrincess at the virtual Burning Man festival in High Fidelity

XaosPrincess (about whom I have written before here and here) has just written an extensive, insightful piece on Medium, titled How to propagate a Virtual World: Conclusions from a dweller’s POV, in which she discusses various problems that she feels led to the downfall of the High Fidelity social VR platform.

As most of you already know, HiFi essentially shut down its operations on January 15th, 2020. While existing users can still sign in, all High Fidelity-hosted domains have closed, and no new user accounts can be created. However, there are at least three forks of HiFi’s distributed, open-source software code currently being worked on by various groups.

Xaos writes:

Having completed my five stages of grief about High Fidelity’s change of direction, I’m now able to view last year’s events through an analytical eye, drawing my conclusions on what can be improved in getting a virtual world to thrive.

Using Mark Zuckerberg’s breakdown of virtual reality (hardware and systems; apps and experiences; and platform services), she lays out several well-reasoned criticisms of HiFi:

In the part most important to Mark Zuckerberg – apps and experiences – High Fidelity both excelled and fell short: While all of the experiences produced by the company…were stellar, apps that facilitate social communication or media consumption – like e.g. text chat or synced video – showed great need of improvement and were often only made possible by efforts of the open source community.

However, she gives high marks to High Fidelity’s platform services:

What always had been High Fidelity’s main enterprise is the second most important area on Mark Zuckerberg’s list: platform services. While quite some users had their doubts about Philip Rosedale’s approach of favoring bleeding edge technology when it came to new implementations, to me personally there couldn’t be any better social VR package than the one High Fidelity was offering:

The open source code enabled community members to implement desired features themselves and is now – that the company has gone offline – keeping its promise of an eternally functioning virtual world.

The peer-to-peer architecture enabled content creators to be the masters of their own domains – without having to follow any TOS, everybody was responsible for their own content – free to install whatever they imagined, ranging from super safe G-rated worlds to X-rated dungeons.

By splitting up the server load into different assignment clients, High Fidelity also managed to gather 500+ avatars in one non-instanced space. Whoever has gone through the hassle of trying to join their friends in an instanced experience or game just has to love this option. And who – like me – also loves the stirring feeling of being part of a large crowd will find nothing comparable in today’s VR environment.

Noting that others such as Theanine had already written at length about the technical problems with High Fidelity, XaosPrincess saved her critique for the social side of HiFi. She raises several good points:

  • that the term metaverse needs to be redefined;
  • the importance of social VR companies in defining, knowing and catering to their target audience (using ENGAGE as an example);
  • HiFi needed to focus on creating a satisfying, bug-free user experience (“Instead of investing into flashy one-time events it might have been advantageous to focus on creating a permanent and entertaining starter experience with a bullet proof tutorial and enticing things to do in order to motivate visitors to come back.”);
  • HiFi didn’t pay attention to competitors and was overly confident that it could replicate the success of Second Life in a different era from 2003 (as I have also written about);
  • High Fidelity’s long history of communication problems with its userbase, which led to a sense of alienation;
  • HiFi’s lack of a clear code of conduct, which left many users feeling insecure: “High Fidelity never stated what kind of offenses would be met with which kind of punishment… nobody could ever be sure how bad behavior would be met.”

But Xaos saved her strongest critiques to one area where she feels High Fidelity made some grievous tactical errors: the company’s impatience to grow. She writes:

Unfortunately High Fidelity applied its mantra “Build it and they will to come,” not only to content creation in VR but also to its real life assets. When there were still no more than 30 concurrent users around in 2018, High Fidelity went on a hiring spree, quadrupling its original headcount to a workforce of 80 while tripling its original office space to two subsidiaries in San Francisco and one in Seattle. I never managed to calculate the exact burn rate, but I believed Philip Rosedale when he argued last April’s pivoting with US$10,000 monthly expenses per user.

To this day this unrealistic growing attempt is inexplicable to me. If I had been an investor I surely would have put a full stop to this amount of spending too, but as one of its highly engaged high-cost users I just wish High Fidelity would have balanced its expenses in line with its slow but steady population growth.

The snowball effect of visitors becoming content creators and enticing new users themselves could have led to an avalanche of attractions for even more new users, if High Fidelity wouldn’t have been impatient.

Instead of giving its community members time to grow into their roles as content creators or event organizers, High Fidelity turned to setting up unsurpassable events by itself, and even paid users for attendance by handing out prizes or Amazon gift cards. This irritated the natural growing process, as these well-meant contests kept the content creators busy, while also being a big competition to individually organized community events.

And when High Fidelity then pulled the plug on all its company operated domains in April 2019, there weren’t enough skilled and motivated community event organizers around to attract new users, and without events there was no more appeal for new users to pay a visit.

As I see it, a more natural growing attempt, incentivizing community members to attract new visitors, could have been beneficial to the overall user numbers.

These are just a few choice quotes from XaosPrincess’ blogpost, which I strongly encourage you to go over to Medium and read in full. Thanks, Xaos!

Liked it? Then please consider supporting Ryan Schultz on Patreon! Even as little as US$1 a month unlocks exclusive patron benefits. Thank you!