Skip to main content

Immersive Payment: Second Update


This is the second update to the Immersive Payment project update under the Grant For Web Fellowship.

You can read the last report here: https://blog.rabimba.com/2021/05/immersive-payment-first-update.html

Background

Technically, WebXR is a device API that makes it possible to distribute VR/AR experiences over a web browser. It is broadly used to describe web based VR/AR experiences.

The web is the biggest platform, able to reach almost every device and balance the power with app stores through permissionless innovation. However, there’s a major issue holding the ecosystem back from growing: monetization.

There’s a reason why 99% of WebXR apps look like prototypes, it’s because indie creators struggle with finding ways to capture value.

More polished experiences like Mozilla Hubs and Frame usually have the backing of a large entity, so they can survive.

While many of the apps on the front page of Oculus Store net millions in sales, apps on the startpage for Oculus browser (#1 app) can barely hit $100 a month in ad rev.

How can other developers and studios justify spending more resources into this open ecosystem like this and make it more than just a hobby?

What Immersive Payment is trying to do

With immersive payment, you are now able to monetize part of the content of a VR scene or asset using Coil. 

Updates:

  • Enabling Integration and First Prototype: The first experiment ensures that VR assets can be run in a WebVR environment. 
    As it turns out it is relatively simple to enable web monetization in a simple webvr page and host it. And it's also fairly easy to add conditional's to the page too. For example to
    <script defer src="cdn.coil.com/monetized-classes.js"></script>
    <style id="wm-stylesheet">
    .wm-if-monetized { display: none; }
    .wm-if-not-monetized { display: none; }
    </style>

    This enables web monetization on those elements.

    Or we can have a whole scene hidden. As you can see in this demo: https://webm-roomscale.glitch.me/
  • Enabling it in a WebAR environment: Using the same concept this can be enabled in an Augmented Reality Environment. The same glitch code detects when it is in an Augmented Reality environment instead of a VR environment and can enable monetization for those elements. 
    • Limitation:
      • This concept is dependent on the coil extension. And for mobile devices even though Firefox is there and has add-on support. It still doesn't support all add-ons, notably, it doesn't support Coil. So any content that is made available part of Augmented Reality will be locked down in mobile devices. Unless we specifically determine the device and unlock that content. Doing that client-side opens up the concept for attacks on desktop devices to unlock content too.
  • How to use it for different contents: Due to the nature of how web pages are built this works on different parts of the elements of a webpage. To enable differential web monetization, the tags should be enabled on those elements. It works for VR assets as well as AR assets.

Open Research Questions Explored:

  1. Can we make the transactions anonymous: This can be achieved by introducing confidential coin concept in the mix. If we want to look at a architectural diagram about how this can be done. This may look like following

Figure 1 provides an example of association between event definitions, event outcome announcements, and event based UTXO notes. It shows nine UTXO notes (named cc1 to cc9) and four transactions (named tx1 to tx4). For each transaction, input notes and output notes are enclosed in a colored bounding box. The coins in front of the arrows are input coins and the coins that the arrows point to are output coins. Each type of data has its own logical chain and the corresponding Merkle hash tree. All the event definitions and event outcome announcements are in public.
They are validated before appended to the ledger. In this work, we assume that event outcomes are declared by the same parties who register the events to the ledger (verified through a secure
digital signature scheme). Verification of the event outcome itself (e.g., status of a tracked shipment) is a separate concern, which could be supported through various established approaches such as
oracle service providers, event validators or zero-knowledge proof based claim verification scheme. We assume that event outcomes are verified as part of the event processing before they
are appended to the shared ledger.
All privacy coins in the system are conditional coins. In the example, one can observe that all the coins (cc1 to cc9) are paired with an event definition. A privacy coin without a real condition is configured to use one of the default constant events as its condition, for instance cc1. This way, all the transactions involving the conditional coins in the system (payments using the conditional coins or transferring of the conditional coins) follow identical validation steps in zero-knowledge protocol when deciding whether the transactions should be accepted by the participating distributed ledger nodes.

This is detailed here: https://dl.acm.org/doi/abs/10.1145/3464298.3493401

Questions

  • How do people pay for WebXR content currently?
  • What is actually being paid for in the present?
    • Access
    • Login / Bot handles ticket ownership
    • Parcels (NFTs)
    • Content (ex : stereo video streaming services)
    • XXX
  • In what ways can we make that process easier and more frictionless?
  • What monetization strategies have yet to be explored?
  • How can artists get paid?
    • Commissions + Projects
    • Patreon
    • Github sponsors
    • Grants
    • VC investment
    • Event tickets
  • Is there a multimodal solution that would work best for in VR payment?
  • Thoughts about non-payment based monetization strategies such as advertising?

Methods:

  • Patreon
    • Many VR creators and sometimes platforms (NeoSVR) are using this
  • Github Sponsors
    • Avaer
  • VC Investment
  • Event Tickets
    • Really unexplored still
  • Discord community with paid roles
    • FIVARS / VRTO did some
    • WHALE / Metafactory using collab.land
  • Virtual market stalls
    • Currently in-progress via Vket5
  • Avatar store
    • Need some sort of VRMchat for people to buy this for webxr
  • Tailors
    • When people buy avatar weawrables they’ll want it custom fit their avatar
  • Virtual land parcels
    • Minting webxr worlds
    • Cryptovoxels
  • Twitch streaming with bits
    • Avaer and Sorskoot are streaming webxr development
  • Engine licensing
    • wonderland
  • App store
    • Construct Arcade
    • FIVARS made a token system for watching movies / games, can be repurposed
  • In-world advertising
    • Elijah from Vhite Rabbit / Construct arcade was exploring with Wonderleap
    • Admix
    • Google Ads work in WebXR
    • No official or unofficial link in how to do that
  • Make it easier to integrate with paypal / crypto / discord
    • Discord path seems easiest
    • FIVARS did a lot of manual work here, could be outsourced?
  • Virtual actors
    • voice actors
    • full body actors
      • performers
    • capturing / minting mocap libraries
      • Train AI npcs
    • Storytelling
      • The Scarecrow: Raindance film festival
      • Live actors in neosvr live performance
  • WebXR can be used for immersive playback

A report of the grant is also available at: https://community.webmonetization.org/rabimba/immersive-web-grant-report-2-28nm

Comments

Post a Comment

Popular posts from this blog

Racecraft (Project Koru) · Prologue — The Origin Story

Racecraft · Prologue , The Origin Story It Started With a Wine List and a Question About Racing How a happy-hour conversation in the Bay Area turned into a trustable AI race coach , and then into a second version that runs entirely on a phone, on the NPU. This is the prologue to a five-part series. Two years ago(1st November, 2024) I was in the Bay Area for a GDE Summit. If you've never been: it's a couple of days of talks among Google Developer Experts, the kind of people who get unreasonably excited about a new on-device runtime, and then , mercifully , a happy hour where everyone stops performing and just eats. We ended up at a restaurant(Puesto Santa Clara), a long table of GDEs, and I was doing the most important engineering of the evening: trying to decide which wine to order. Across the table was Ajeet Mirwani . I don't even remember how the wine talk turned into racing talk , these things drift , but the moment the word "racing" ...
Post a Comment
Read more

The Throughput Trap: Benchmarking vLLM on OpenXLA and the Reality of Production LLM Serving

vLLM Systems · DevLab 2026, Deep Dive I was recently invited by the Google TPU team to speak at the OpenXLA Summer DevLab 2026 . This post breaks down our deep-dive evaluation of the matured vLLM + OpenXLA stack, the fundamental engineering mismatches between CUDA and XLA serving paths, and why traditional capacity metrics are lying to you. If you are operating large language models at enterprise scale right now, your platform architecture team is likely staring at a massive infrastructure crossroads: Should we migrate our core serving workloads from GPUs to TPUs? Historically, NVIDIA's CUDA ecosystem was the only serious option for user-facing, low-latency LLM generation. But here in 2026, the economics and infrastructure options have transformed. Google TPUs are highly available, cheaper per chip, and the open-source serving stack built around vLLM and OpenXLA has officially achieved absolute production readiness. Yet, when our infrast...
Post a Comment
Read more

A Split‑Brain Neuro‑Symbolic Training Method for High‑Velocity Autonomous Coaching from Telemetry

 Author: Rabimba Karanjai Scope: Problem statement + data methodology + model training (no deployment discussion) Abstract Real‑time coaching in motorsport is a safety‑critical learning problem : a system must map noisy, high‑frequency telemetry to short, actionable guidance that remains physically consistent and avoids hazardous recommendations . This paper proposes a “Split‑Brain” training formulation that separates (i) a semantic coaching target (what action/critique should be expressed) from (ii) a reflexive interface (how actions are represented as compact, verifiable tokens). The approach trains a Small Language Model (SLM) in the Gemma family [1] using QLoRA fine‑tuning [2] , and introduces a telemetry tokenizer plus teacher‑student synthesis pipeline to generate instruction‑action pairs at scale. Core contribution: a reproducible method to convert “ golden lap ” differential tel...
Post a Comment
Read more