[Opensim-users] Hoping for a fearless comparison of opensim vs unity 3D

[Justin Clark-Casey]

I think that OpenSimulator and Unity have some overlap but not by a huge amount.

My perspective is that the focus of Unity is very much on game development.  It gives you a good and flexible set of 
tools but you need to do a fair amount of work to plug them together or extend them to create a high fidelity (ha) 
product.  The focus is on creating a one-off experience, though the lines are blurring now that some games (e.g. 
Minecraft, DOTA2) are very long lived and keep receiving updates.  The experiences are high quality because they are 
quite tightly controlled.  High multi-user (let alone massive multi-user) has not been a focus area because this stuff 
is *hard* and nowadays not obviously a winning formula for gamers.

For OpenSimulator, the focus and much of the raison d'etre is the unified and persistent virtual world.  Thus, it gives 
you a high level set of tools which are much less flexible (inventory, attachments, linksets, etc.) but because everyone 
has them it allows collaboration and content reuse at a high level (e.g. scripted objects, OARs).  Some games blur into 
this (Minecraft, etc.).  It's a free-form environment so there's a high degree of freedom but a lot that can go wrong 
(analogous to open-world jank) [1].  I see it as more web-like because the same high-level software is evolved over time 
with the hosted content changing.

Moreover, there's a very high social focus through time.  Because the same high-level concepts are shared, there's more 
scope for network effects (esp. with the Hypergrid) but the technological base is much more primitive and relatively 
unexplored.

So whilst I think Unity makes sense in many use cases, OpenSimulator is ultimately much more interesting to me 
(unsurprisingly) because it gives a glimpse into something radically new, a distributed, anarchic and evolving Metaverse 
rather than a single vendor game.

I think there is vast scope for the OpenSimulator ecosystem to continue to evolve with features such as template 
objects, multi-level linksets, more intuitive viewers and to adapt to technological evolution as embodied by new 
hardware such as the Oculus Rift.  Because it's open-source, innovation can happen anywhere and without a single 
company's permission.  I believe the critical thing is that we arrive at protocols and formats that allow evolution by 
disconnected parties whilst still inter-operating with the existing system.  Again, it's a comparison with a web 
ecosystem that has extensible formats such as HTTP and HTML (insert a tag that a browser doesn't understand and it 
doesn't (usually) stop your whole page from rendering).

However, arriving at these formats and solving other hard fundamental problems takes an enormous amount of time and 
effort, not only through writing code but also in discussion and co-operation between parties with different interests. 
  My hope has always been that the platform will become interesting enough to attract the critical mass of academics, 
enthusiasts and entrepreneurs who can generate the time and funding required.  To some extent this happened but not 
enough (as of yet) to win any significant attention outside of this niche.

[1] http://www.giantbomb.com/open-world/3015-207/

On 21/07/14 16:43, Wade wrote:
> This discussion has been the most enlightening  I've seen in a long time!
> Thank you everyone!
>
> My experience agrees that faculty don't generally want to learn 3D content creation.
>
> Students are an interesting mix, and in high-stress programs also have very little tolerance or capacity for steep
> learning curves.
> ===
> *On simplicity *
>
> In terms of students building things that didn't exist,   maybe there is a game-principle based sweet-spot,  because
> it's clear from the numbers that tens of millions of people spend tens or hundreds of hours with Minecraft.
>
> That suggest to me that students would love to co-create cool stuff, but the interface for doing so needs to have an
> extremely extremely simple /*starter subset*/.   I say "starter", because gaming-principles also show that people who
> stick around and pay for worlds like World of Warcraft*_like challenges_*, or "unnecessary difficulties" as Jane
> McGonigal's "/*Reality is Broken*/ - why Games make us Better and How they can Change the World" book explains so well.
> (Imagine the interest in golf if the average length from tee to hole was ten feet, in a straight line, on a flat course,
> and the hole was ten feet across.)    This is a great book, by the way, and very eye opening and challenging a lot of
> misunderstood concepts about "games", the nature and type of feedback that works,  and why so many people voluntarily
> spend so much time on them, that is directly applicable to building any learning environment.
>
> For experienced builders (or those past their anxiety - resistance stage), yeah,  prefabs in Unity are great!
>
> What is even better is that in Unity you CAN build/*hierarchical objects,*/  then mix and match the parts.  In OpenSim
> and Second LIfe,  once you put the wheels on the car and make a link-set,  all traces of "wheel" are gone, and it
> becomes absurdly difficult to go back and put different wheels on the car if each wheel has 47 parts like spokes or
> lugnuts.     You can approximate some of that capacity with "Builder's Buddy" or other tools that let you rez an entire
> multiobject scene with one click, but those are a true pain to load and maintain.
>
> So,  whether it's Unity or OpenSim,  I think one thing that is needed that is very hard to still see for Virtual reality
> natives is exactly HOW SIMPLE the INITIAL interface has to be, so that it is satisfying and rewarding to try to use for
> a terrified newbie, peeking though the fingers of the hands over the eyes.   So simple in fact that even a faculty
> member might say "Oh heck, even I can do THAT!".
>
> ===
> *On "weakest links" in collaborative environments*
>
> And both faculty and students are greatly upset by technological failure where they are used to trivial behavior, such
> as having voice working.   The collaborative environment is much harsher than individual user environment since for
> voice (or many other things) to actually be useful,  it has to work for EVERYONE, not just most people.
>
> This is a feature of collaborative environments that I didn't realize till Gary Olsen pointed it out.  A collaborative
> environment can become a "weakest link" exposer, where everyone's experience is limited by the least capable user.
> This is one of the issues with, say, Electronic Health Records systems that is underappreciated and distinguishes it
> from, say,  a cloud-based spreadsheet.
>
>
>
>
>
>
> _______________________________________________
> Opensim-users mailing list
> Opensim-users at opensimulator.org
> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-users
>


-- 
Justin Clark-Casey (justincc)
OSVW Consulting
http://justincc.org
http://twitter.com/justincc