5 ms·
As someone who's been part of a number of communities in SecondLife for 15+ years, this is exactly the problem that platform has had in becoming THE metaverse.
by g5095 5y ago
As someone who's been part of a number of communities in SecondLife for 15+ years, this is exactly the problem that platform has had in becoming THE metaverse.
Despite technically having everything required to tick that box, development is shackled to the funding capability of a single company, land values are extortionary considering the cost of server hosting these days ($300+ for a 'sim' a month, when they shard them up 100s to a server).
The solution has to be a set of standard protocols and interfaces that can be implemented by anyone, just as 'the web' was. Every time I see a 'Secondlife killer' come along I judge it's success at taking out the incumbent on it's openness, and I'm yet to see a real contender.
- cblconfederate 5y agoOpensimulator exists
- Animats 5y agoActually, each full region (256m x 256 m) uses about one CPU and costs $350/month. Hundreds to one server, no. 16 on a 16 CPU server, yes. The architecture is rather expensive to run. Operating cost is per region, not per user. Regions run 24/7 regardless of whether any user is in them. Trains continue to run, crops continue to grow, grazing animals graze, a few cars drive themselves around, traffic lights continue to change... Improbable's Spatial OS tried a more flexible architecture, where regions were dynamically resized as users entered and left. It cost US$400 million to develop, was expensive to run, and all four games based on it shut down due to high costs. A big seamless virtual world is a hard problem. Except for the voxel crowd (Roblox, Minecraft, Dual Universe) who seem to be good at scaling, at the cost of being blocky.
- mojomark 5y agoThis argument/counterargument is pretty eye-opening to me. Kudos to both Animast and g5095 for the discussion. I work in a completely different domain than gaming, but there is no question that this strikes the nerve of a universal challenge that we have continued face as humans seeking to better understand the very fabric of nature and civilization itself. Please don't take offence, but g5095's stance is akin to early US federalists advocating for strong centralized governance, whereas Animats stance is akin to US democtratic-republican values of distributed governance.[1] If you think about Animast's point, that "regions run 24/7 regardless of whether any user [citizen] is in [consciously using] them. Trains continue to run, crops continue to grow, grazing animals graze...", and the the discrete energy cost of operating those regions - is this not the very challenge that all governments face in the "real world"? With the maturation of the 'Metaverse', in which all functions of nature are boiled down to the discrete (countable) energy packets needed to transfer information, are we not seeing the struggle between centralized vs. distributed government play out in virtual space? If so, can we leverage the Metaverse as a proxy for nature, generally, to find the mathematical ideal balance between these two competing forces of democratic civilization? The result, I would presume, being some difinitive proportion - the measurable distribution and magnitude of 'control' energy with respect to the distribution and magnitude of total energy of a predominantly nonstochastic system (society). It seems like there is a nexus here between control theory, information theory, thermodynamics, and indeed, sociology, that can be studied using the Metaverse to arrive at a more balanced societal framework. There is a truth here begging to be understood. 1. https://www.ushistory.org/Us/19c.asp https://www.ushistory.org/Us/19c.asp