Have you considered a proprietary building system instead of becoming a cheap mesh rendering service?


#1

This is a question directly targeted towards the developers. My question is as the title of this topic describes, have you considered a proprietary building system instead of becoming a cheap mesh rendering service?

My reasoning behind this is when I look at this platform (I have been looking) I do not see much value in it when everybody can link a mesh they find on the internet.

Why would any content creator or artist spend days or weeks building quality items for a small userbase, when they have to directly compete with every possible model that is availabe online? Your company would like to make money from the marketplace you have, why would a user buy a furniture set from your marketplace when they can grab a furniture set for free from the internet?

Next to the financial incentive your platform might offer to certain individuals there is also the creative aspect. A content creator might work 10 hours per day in Blender and spend 3 minutes per day in High Fidelity. By being forced using proprietary building tools and keeping everything inside of your platform content creators would also take up their role as active participants in the world compared to be outside merchants.

My next point is that for 99.7% of users and visitors using Blender or other 3D model software is out of the question for various reasons, creating in Blender is work, tedious; lonely; repetitive and dull. After a day of work at the office a person will not feel like launching Blender and spend some hours extruding faces by himself. People might still want to log in to their domain, chatting with their friends while doing some building work on their zone.

Another point I would like to make and perhaps the most important one so far is this. With the introduction of sculpted prims in Second Life this did create two classes, namely consumers and creators. Before the sculpted prim was introduced anyone could participate, the tools were simple enough for 85% of the population to understand and use. People could modify their purchases, create together. Instead of lonely extrusion hours in Blender creators were community members and did contribute to the world.

There is a platform named the VWF or virtual world framework which functions just the same like High Fidelity in terms of content upload. There is not much interest for anonymous mesh uploading services from what I can tell.

I am looking forward to hear your views regarding this, you are still in early alpha now and able to go any direction you want.

  • The proprietary tools I speak about would not be simplistic prim type or block type building tools, such a tool would allow fabrication of advanced mesh models depending on the actual skills of the user. The better your skills get the better your models will look and the higher the value of the model. A set of tools allowing the complete newbie to create his first chair without much study, even more simple compared to using Google sketchup.

#2

I have no interest in using a propriety building solution aimed at the lowest common denominator.I think that using common 3d formats is good because these are used both by hobbyist’s and the industry. This means when were messing around with these tools were actually learning a trade.
I don’t find Blender repetitive tedious or dull.
I do like the idea of communal building but if i can only do it using an immature propriety format then again I wouldn’t bother. I think building blender into the interface so people can communally build would be cool but i don’t know if that’s even possible.
I think throwing in prims to make a dated version of second life or trying to be mine craft would be foolish. These platforms do their thing very well.


#3

If you want proprietary, there are already Convrge, VRChat, AltspaceVR, etc. etc. (Incidentally none of them have a linux version). It’s my understanding that hi-fi is the counter-design to those tools: Open Source and inclusive.

By being forced using proprietary building tools and keeping everything inside of your platform

I hope that people are going to wise up and realize that these things may make money in the short torm, but in the long term proprietary and restricted platforms are bad because by embracing them you handed over control over what you do with your content to a corporation.

My next point is that for 99.7% of users and visitors using Blender or other 3D model software is out of the question for various reasons, creating in Blender is work, tedious; lonely; repetitive and dull. After a day of work at the office a person will not feel like launching Blender and spend some hours extruding faces by himself. People might still want to log in to their domain, chatting with their friends while doing some building work on their zone.

High Fidelity is open source. Blender is open source. Someone will undoubtedly integrate a VR UI for blender into High Fidelity - once there is demand for it. In my opinion it all hinges on the thought that more people are going to see the value in openness in the near future.


#4

Do you mind telling us where you found that metric?

While I disagree with the above (because I think you just made up some number) I DO agree with your other point,

As I have struggled with this myself. I know Blender. I know it fairly well at this point. But it does become a bit of a task to log into my machine after a long day at my ‘real’ job. When HiFi matures, and the distributed computing problem is solved, this platform will hold much more of my time and interest. For now, it’s really interesting to see things evolve.


#5

Blender is only tedious repetitive and dull 7% of the time its at least 40% infuriating. 25 % of the infuriation is caused my my own incompetence/stupidity with lets say 16%smugness when something works properly and the remaining 9% was a made up statistic .


#6

I think HiFi does need building tools if only to help prevent it from being seen as a regular 3D engine like Unity3D. Although I doubt that many people would use the tools in the long term because casual 3D modeling solutions tend not to be very flexible or optimal when it comes to poly counts etc.

It looks like HiFi wants to embrace HMDs and so maybe a fun modeling solution can be created with HMDs in mind. So instead of an interface built around the mouse and keyboard, the interface could be built around the 3D handheld controllers. I imagine the controllers would be a little clumsy to model with but people may find them entertaining for a short while. But ultimately the quality of the models created with such tools would be pretty awful.

But any attempt to create full featured modeling tools in HiFi would be the equivalent of reinventing the wheel. It would take years to create decent modeling tools and you would still be left wanting more features. And then there’s the question of what would those tools look like?. Personally I’ve tried all types of modeling over the years, polys, voxels, nurbs etc. Poly modeling is still the most efficient and reliable method by far.


#7

The origins of … “the Virtual World Framework (VWF), which is being developed by defence giant Lockheed Martin and funded by the Office of the Secretary of Defense. Lockheed was awarded a contract to develop the VWF by the Department of Defense in January 2011” (ref http://www.army-technology.com/features/featurevirtual-world-framework-the-next-collaborative-training-revolution-4445859/) doesn’t entice me in the least. Another lock in and run by the DoD haha you have to be joking - no no, hell no kick out the jams brothers and sisters


#8

AlphaversionD I intended to write a lower percentage like 90% but when thinking about it 99.7% is 997 out of 1000 non 3D model creators or visitors. Out of 1000 users (Second Life users I did base this on) you have 3 who produce mesh content using programs such as Blender. There are 40 million Second Life accounts this makes 0.3% or 120000 mesh creators which I would describe as pretty decent estimate.

Several people mention they want industry standard tools which also means you need to handle all the downsides that come with the industry. I am also not speaking about low quality producing tools, I speak about tools or a system that allows you to produce objects very close if not exactly how you can build in Blender if you get good enough, but even the absolute newbie can also make use of the tools to get started.

These are design choices High Fidelity needs to make, industry tools might look interesting, easy to get a start but they come with a whole lot of baggage.


#9

The main thing that attracts me to HiFi as a virtual world creation project, vs the other virtual world creation projects, is that it’s designed from the ground up to be decentralized and to NOT have all your assets and credentials stored on, and thus “owned by,” one corporation. Now, true, having a designed-specially-for-HiFi mesh creation interface and mesh format or something MIGHT seem attractive… but with the whole system being meant to be opensource, that wouldn’t stop someone else from simply making an OBJ-to-HiFi-mesh and Collada-to-Hifi-mesh converter routine, and you’re right back to having every 3D object you can find online be importable to HiFi world again.

Anyway, I don’t want something that is going to tilt HiFi back over towards the walled-garden approach that I came here to frakking get AWAY FROM.


#10

I disagree with your points fon Blender.

@judas made most of my points, but here are some extras:

  • avatars, cant just grab any just any model, you have to work to convert them. And work ever More to get shapekeys to work as you want.
  • ability to use 3d files off internet can be blocked by the provider after enough traffic bleeds through hifi agent
  • hosting the models your self requires you to pay for the bandwidth. Also opens you as the host up to dmca from unhappy content creators.

Models seriously are no small things compared to images. They are bandwidth hogs, especially of you start having 20-30 downloads a day. Talk about in 1000 and you got issues (3 mb static model + 2 mb textures, easily hits 5 gb a day. With 100 gb bandwidth server be budget, it is noticeable.) This is supposed to be solved through the p2p nature of ATP where assets are distributed shared across domains if the asset is allowed to do so…

  • hosting through the ATP protocol stuff its not as easy distributeas is: since its not just a url, instead an uuid allocation.

On the side however, id like to remind you that there is indeed a modeling tool being worked on…
Which is through polyvox. As of the moment its still in development though, but it does provide all the simple requirements .


#11

Well, my understanding is that the individual domains at some point will have what amounts to an own, local asset server for all their rezzed content, sometimes spreading the processor-power required for handling the rendering of that stuff across other, selected, “nearby” domains. The situation of having to load the model from a separate system such as a cloud storage account or someone’s FTP server (i.e. EVERY copy of a particular chair that one might see rezzed on hundreds of domains gets loaded by each visiting user off of, say, turbosquid or something) is more of a stopgap / placeholder until this other system gets far enough along to be incorporated into the domain server mechanism.


#12

I like the open concept of hifi. The online editor, which we have at the moment, is a script. So every programmer can change it and add tools to it, how they are needed. This may not go from one to the other day. But so much has developed in the last month.
So maybe a priority list of additions from user view would be helpfully.


#13

I think this is a large issue worth considering, maybe with a wider perspective than stated.

I still don’t really get what the HF business model is, or even if there is one, or even what HF will look like in beta or production when those days come. But taking the view of someone who would use the HF infrastructure to build businesses, potential business models are the basic thing. How and by whom worlds are built, what options users of different skill levels and motivations have to build, how products can be controlled by permissions and copyrights protected, those are all critical.

My take is that HF, as an infrastructure, must provide businesses or even hobbyists or organizations that can afford to create and maintain domains, with maximum flexibility in all those respects. For building, we can’t expect all users to have the talent, time and willingness to learn and use the pro tools. The relatively few that do will want protection of their work and control over how and who can use it. In-world modification functions, especially for avatars would seem to be a basic and obvious function to provide. Same for other building and terrain modification.

I don’t know if that can be done entirely on the application (domain) level or if it needs support in the infrastructure. I’m still foggy about this division between infrastructure and what is built with it.


#14

Is that actually in the roadmap?


#15

There are a lot of exiting ways to create 3D models these days …

Blender or maya are great ways to create 3D content.
Also new ways are coming up, like you can make 3D models with 3d cams or even with your phone.

Hifi will get some inworld build tools too. The terrain will probably be editable with inworld tools. Some javascript tools to create inworld mesh will also pop up over time.

And yes, if someone does not have got the time to learn the tools to create the 3D models and would rather socialize with his/her friends … well then I do I not see the issue. That person will just not create models …


#16

I think Blender, like a cold glass of beer, is peepee Dickie (apologies to Putney Swope)


#17

That domains-will-host-their-own-objects thing was described in one of the Friday meetings a couple weeks back. At least, that was the understanding I got of it when I heard it described anyway. It’s possible I misconstrued something, though.


#18

That would make sense, and would seem like something domain owners would prefer for some degree of control. It would be nice to have flexibility, though. For users who want to ‘own’ their avatars and bought or self-created stuff and keep them as they wander the met averse, a private store of some kind would make sense. Sorry if this is pulling off-topic.


#19

Specifically @Nathan_Adored is referring to the ATP protocol I mentioned (Asset Transfer Protocol)