Comment by mintplant | Hacker News Reader

Comment by mintplant | Hacker News Reader

Parent Full thread

I can't help but compare the metaverse-y projects coming out of the VR world with Second Life. For more than a decade, it's had:

- an interconnected world of user-built locations, with near-seamless travel between them

- an environment in which every object is user-crafted and scripted, and can interact with each other and with users in a dynamic networked physics simulation

- a client which bundles the development environment, including 3D modeling and scripting tools, fluidly into the main game interface, such that objects can be created and edited on-the-fly and experimentation can happen live alongside other users in a social environment

- its own currency, convertible to real-world money and back via an actual exchange, and tools for conducting commerce both in-world and on the web via an Amazon-like marketplace

- infinitely-customizable avatars which allow for combining components from multiple sources and creators, spawning a micro-niche of virtual "fashion" trends, events, and blogs

What holds SL back is the accumulated cruft of ~1.5 decades of development mistakes [0], legacy technologies, and backwards-compatibility requirements with an ocean of content. But it's always been worlds more ambitious than modern claimants for the title of 3D social environment of the future, which seem content to be multi-user VR YouTube clients.

[0] I could go on and on about the mind-boggling pile of hacks that keeps SL ticking, but that'd probably drag this thread too far off topic :)

Though I'm not familiar with SL, I'm pretty familiar with MMORPGs and similar software. I think what set the late nineties and early 00's projects apart from what is being done today was that they were pure passion projects. 1 to 6 individuals in remote basements just grinding it out over the course of years.

Nowadays a lot of big VR and AR efforts are funded by millions of dollars, the founders who should have the passion aren't doing the actual development, and many if not all staff are probably anxious to leave and catch an uber for $12 cocktails at the local hotspot.

That's not a bad thing. God forbid it's healthier. But it is a different environment and we can't expect the same creative output.

As one of the people who founded one of those projects in the ought's I can say a couple of things. There were only two in the early 2000s, SecondLife which raised 11 million in 2000, and my company Prototerra which raised 500k. SL came out to much fan-fair (amazing what a little PR will do for you). While we struggled to get any press. Eventually we did well being the tech behind a number of money making sites. Around 2005 there were quite a few new ones that raised a lot of money, none still exist. Neither does Prototerra, crash of 2008 killed business and the 3 guys working in the backroom had families by then, but it was fun!

OnLive Traveler came much earlier

My friend worked on it back in the 90s told me about it. It was licensed to MTV. Huge cult following.

I found this: https://archive.org/details/vw_traveler-bill_gates

Indeed there was a number of previous generation virtual worlds using VRML. Lets say they left a lot to be desired.

There were also all the blaxxun-based ones, such as Cybertown or the one from Canal+.

I'm assuming you've heard of Sansar, the VR project from Linden Lab. I haven't tried it, but presumably they are incorporating a number of lessons learned from Second Life.

https://www.engadget.com/2017/07/31/sansar-vr-creator-beta-s...

Sansar is nice, but not directly comparable to SL (it reminds me more of would-be SL competitor Blue Mars). There is no single world; instead, it's a platform for the creation of stand-alone "experiences" (which can include teleport gates to other experiences, by all means). The building tools are also limited. Judging by my (so far also limited) experimentation, you end up working alone, mainly in something like Blender or Maya, then you upload and position your meshes in Sansar. In SL, you can collaborate on builds in real time.

An interesting open source alternative in the social VR space is Phil Rosedale's High Fidelity: https://highfidelity.com/

But personally, I am currently infatuated with the much more light-weight Rec Room: https://www.againstgrav.com/rec-room/

Rec Room is amazing! really solid experience ... could easily be expanded upon to really capitalize on this idea of a shared virtual world IMO

I've played with High Fidelity. It constantly disconnects, and it can take over a minute for an area to load enough to provide basics like 'a floor to walk on' or 'the shapes of buildings'. Second Life ran a hell of a lot better on my crappy old computer on a crappy Internet connection; the graphical improvements of High Fidelity are scarcely better and certainly don't justify the poor performance.

Oh, and it's already full of furries, so you might as well torch it and start again.

> Oh, and it's already full of furries, so you might as well torch it and start again.

Many of the most prolific creators I knew on SL had some affiliation with the furry fandom. I would take that as a good sign, actually: they tend to be early adopters.

> It constantly disconnects

For what it's worth, I've never had that problem.

Does it also happen with your local sandbox?

Nope! My Internet connection is less than perfect but better than the national average.

The local sandbox is still laggy crap on a recently-built i7 desktop with a 1-year-old NVidia card, though.

JanusVR is worth looking into. The markup is an HTML like language and it has browser support. https://JanusVR.com

I also think that SecondLife stands on its own as far as scale and ambition. And although I can imagine it being engineered better, I can't point to any other project that is similar enough to use as a legitimate example of better quality.

In a way it almost seems like the inverse of Craigslist. SecondLife feels so over-the-top and sloppy that I think someone will come along with some new technologies and replace it with a leaner more performant product. But they haven't.

Its a rich mans hobby. Just expensive as hell to develop. Art assets cost a small fortune. Anyone remember Kaneva? They are still dropping 20 million a year into it with nothing to show, and that's been over 10 years. Helps to be a billionaire!

opensimulator is like second life, but free

> [0] I could go on and on about the mind-boggling pile of hacks that keeps SL ticking, but that'd probably drag this thread too far off topic :)

Perhaps, but that's the sort of diversion worth reading. Please continue.

I'd like to hear about this too :)

Just to begin with: there is a legacy avatar system that half the people use. It has clothing system etc. but it's ugly, so other people hide their legacy avatar with "alpha masks" and wear "mesh avatars", which are meshes rigged to the same armature as the "legacy" avi. Then there are different rigging methods, the avi has a set of basic legacy bones, a set of volume bones, and they recently added another set of bones. Their market contains clothing and skins/meshes for all possible combinations of those hacks. What's even more confusing is that SL gives by default a "mesh avatar" to new users, and the poor guys have to learn about all this clusterfuck after the fact.

There is also opensim , the open source "second life server", and thousands of smaller grids . Also, highfidelity.io is aiming for a VR version of second life (like Sansar) but is still in development. Also, sinewave.space wants to make a modern second life.

Reply on news.ycombinator.com