We Will All Pay For The Metaverse

Will the Metaverse bring us all together, or will it further separate us (who have immersive access) from them (who are not hyperconnected) and deepen the digital divide?

The Internet's best effort delivery model does not specify any minimum bandwidth or availability or maximum latency (delay) between the sending and receipt of a datagram. The federal government defines broadband in terms of minimum upload and download speeds, but those numbers are perennially criticized as too low. A goal commonly cited by broadband activists is 100Mbps symmetrically. But the true determination of the adequacy of Internet connectivity is made by end users of popular applications. The operators of services such as Netflix, Zoom and online games publish their own minimum requirements to enable a satisfactory user experience. Application requirements are the drivers and the ultimate arbiters of adequate broadband connectivity.

In the telephone market, local and long distance services were traditionally metered and billed at different rates. As the cost of domestic long distance service fell, it is now only international calls that are billed separately. Common international destinations are sometimes combined into tiers of service to which unlimited access is provided for a fixed additional fee. The emergence of an Internet that can (sometimes) support audio and video communication using unmetered worldwide service has led to a segment of the teleconferencing market being provided to end users at no cost beyond that of their Internet connection.

It is important to note that the cost of Internet teleconferencing is not zero. There are two major categories of teleconferencing services: Cloud based and peer-to-peer. Cloud based services, which are the most common today, require the operation of dedicated or virtual servers located in regional data centers. The cost of these servers is either covered by the subscription fees charged to the (mainly business) users of premium teleconferencing plans or out of the capital investment of investors. Although peer-to-peer teleconferencing can avoid much of the cost of Cloud servers, all teleconferencing relies on the existence of an Internet which enables high bandwidth and low latency connections either to the Cloud or directly to other end users. The cost of this high quality connectivity is part of the overall costs of Internet Service Providers (ISPs).

As with long distance telephone calls, there is an argument that folding some of the cost of teleconferencing into the cost of generic Internet connectivity makes sense. There may be little or no incentive to the ISP in offering a lower grade of service that does not support teleconferencing if most end users are willing and able to pay for it. While it is possible to argue against this policy, we will accept it for now. After all, everyone needs unlimited videoconferencing and no one has any issues with their Internet connectivity costing too much, right?

"In a recent?article in?LightReading, Mike Dano quotes Dan Rampton of Meta as saying that the immersive metaverse experience is going to require a customer latency between 10 and 20 milliseconds." - Doug Dawson's Pot's and Pans blog

If immersive connections to the metaverse will require extremely low latency guarantees then either the servers which implement the metaverse will have to be very close to every end user or else the access network will have to be highly resourced and very well engineered and operated. There may very well be limits to how many data centers we can build and where they can be placed, not least among them being energy consumption. So it is all but inevitable that the immersive experience will require very expensive low latency connectivity in the "last mile" connection between an end user and their nearest metaverse point of presence.

In the current one-size-fits-all model of Internet broadband connectivity (where maximum last mile bandwidth is the only resource that is billed differentially), the high cost of extremely low latency will be spread across the connectivity fees paid by all end users. It is becoming common rhetoric that very low latency is simply a requirement of high quality Internet connectivity. If it drives up the true cost of everyone's Internet, we should all just demand higher subsidies or, if we don't qualify for those subsidies, just grin and bear the cost. You wouldn't want to be the only one at your workplace who doesn't spend all their leasure time immersed in the metaverse. What would you talk about around the water fountain?

I am not so foolish as to predict the success or failure of the metaverse as the inevitable evolution of personal and business Cloud services. However it is worth noting that a three-dimensional version of the Web based on the Virtual Reality Markup Language was proposed over 20 years ago but never caught on. Various forms of augmented and virtual reality appliances and systems have been available and gained traction within the gaming community but there are challenges in the technology's practical use as a more general form of social media and remote collaboration. Apple and Meta both entering the market is no guarantee. Not every logical extension media to a new modality is successful - we should all remember Smell-o-Vision!

What can be predicted is that accepting stringent latency requirements as part of the definition of general broadband connectivity will have the effect of raising the true cost of such connectivity and making it less accessible to those currently on the wrong side of the digital divide. If, like extreme high bandwidth, we accept that extreme low latency should be billed differentially in order to allow us to continue moving closer to universal access, then there is a question whether the latencies required current broadband Internet services are standing in the way of that goal. If, as the United Nations declared in 2016, Internet connectivity is a basic human right, necessary for all commerce and civic engagement, shouldn't we be seeking to deploy it in a form that could be truly affordable to everyone on Earth or even free when used for those purposes?

Further reading on the challenge of universal Internet access: Is Universal Broadband Service Impossible?, Micah Beck, Terry Moore, arXiv:2204.11300, April 2022