Friday, September 23, 2016

2016 OSHG: OpenSim - What it will take to get into prime time? By Steve LaVigne. HGProjects

OpenSim - What it will take to get into prime time?  
Invited post by Steve LaVigne,  CFO A Dimension Beyond, Inc.
As part of the OSHG initiative, I am posting articles by members of the Organization Committee and potential members of the OpenSim Development Committee.  All articles are open for comment.  The intent for these committees is that they represent the interests of the whole OpenSim community.
  • Steve LaVigne is Halcyon-Moses Liaison on the OSHG Organization Committee.
  • Steve Franklin is the av name for Steve LaVigne
  • Emphasis (bold) by Selby Evans.
  • The views expressed are those of Steve LaVigne.

Multiple ways to use OpenSim

  • Keep in mind that there are multiple ways to use OpenSim
  • As in SL: Register on a grid, load a viewer, enter and look around
  • As in SL: Rent land if you want
  • Hypergrid: Enter a hypergrid-connected grid, go to other such grids
  • Run your own grid on a grid host: INDEPENDING HOSTING
  • Run a grid on your own computer: Sim-on-a-stick

Editor's note

  • When I (Selby Evans) publish other people's work, I shift from author to editor.
  • I will only change the layout a bit, add hyperlinks, and make minor changes for clarity.  
  • I post the article because I think some of my readers will be interested in it.
  • The views expressed are not necessarily mine.  
  • You can reply to the author in comments on the article.  
  • No captcha, no sign in, no delay in you reply promptly.


  • 1. Hardware issues:
  • 2. Software issues:
  • 3. People and attitude Issues:
  • 4. Avatar issues.
  • 5. OpenSim needs to be monetizable!
  • 6. OpenSim and the handicapped.
  • (More after the break)
  • Where is Arcadia?

Metaverse events, current and upcoming


1. Hardware issues:

A. In order to get OpenSim more widely adopted it needs to be easily installable and
maintainable. This means having it already set up and configured as customers need it.
Could be VM (Virtual MAchine)  or a whole server set up and ready to use. MIAB was a good example of this. This can be handled by a competent service company.
B. The viewer side hardware at this point is expensive to purchase as OpenSim tends to
need a lot of resources to operate well. As hardware graphics capabilities increase
OpenSim will run better in more devices. OpenSim needs to be able to run on most
anything including cell phones, Chromebooks and tablets. The problem is that most Government and corporate equipment is built to a lowest bid making hardware not so good at thevery beginning.

C. Have a standard set of specs that goes into detail what hardware is needed for what
performance levels. Also further testing needs to be done with virtual machines and
containers like Docker.


2. Software issues:

A. On the viewer side, OpenSim needs to be able to run in a browser. This browser
viewer needs to have a simple, easy to understand interface that even the most
clueless person can understand and use. There need to be several levels of viewer
ability levels from noob to expert builder to Admin.
B. The viewer should be just a web page like any other. Going to different grids might
not need the current hypergrid teleport system. If a persistence of avatar system exists,
then an avatar could go from one grid to another easily by going to another link on a
webpage or an object in-world. This needs to be secure and and encrypted. It also
needs to have secure video, chat, and voice communications.
C. When a person initially logs on there should be no need to set up an account on the
system. The system should be able to log you in and give you a basic random avatar
without you doing anything but logging into the web page. If you want to have a custom
name and avatar there needs to be an easy way to sign up for one and have that
identity and look go with you wherever you go.
D. Database issues. The current way the database is set up leaves a lot to be desired.
For one thing the size of the DB tends to get very large very quickly. Databases can be
several TB in size on a bigger grid. This leads to corruption of the DB and errors which
eventually cripple the system. At this time OpenSim does not really clean up it’s DB and
eliminate duplicates very well. Some advances have been made in this area, but not
nearly enough. WhiteCore, another version of OpenSim has been experimenting with a
file, or URL based version of a DB that they say is faster. This might be a good thing to
look into as they say that in their testing it is faster. Another benefit would be simpler
backup and more security if it is done right. A File based database would also keep the
main database small as it would contain only text pointers to file locations. There also
needs to be a system set up that eliminates duplicates in the DB. We also might look
into other systems like Mongo DB, Hadoop, and such.
E. Easy control and branding. MOSES created an interface that controls their version of
OpenSim. It is called MGM or MOSES Grid Manager. Michael did a really good job on
this but some things got left out. MGM needs an easy way to set up the splash page
and login page as well as to be able to rebrand these things. As it is currently you have
to dig around in the code and figure out what is what and change it at the code level to
modify anything on the interface. A web page interface for admins would be a great
improvement and lead to wider adoption. All functions of OpenSim need to be
controllable from a simple brandable, easy to use web page interface.
F. Clean up and annotate the code! OpenSim is about 10 years old now. It is a patchy
mess in lots of ways. It needs to be cleaned up and annotated so it can be easily
improved in the future. All functions need to be modular with a core that everything
works on like an operating system.
G. Get a good Web server like NGINX in there. The current web server is way out of
date and not up to the needs of a secure Virtual World system. Security is still an issue
with vulnerabilities in the database and other places.
H. OpenSim needs a downloadable viewer that is simple to use and pre-configurable
easily. Currently some people have taken viewers like Singularity and Firestorm and
modded them to their own needs. If we are going to use a regular downloaded viewer it
needs to be easily configurable, brand-able and skinable. This needs to be able to be
accomplished by a built-in system that makes it easy to use. Currently there is some
movement in this direction, but we need to go way further. In some ways this is a
stopgap product or a specialty product that would largely replaced by the Browser-
Viewer. I do think that creative people building in OpenSim will want this product and all
it can do. Some companies and schools might want a specialty viewer for their people
I. A secure built in system for chat, voice and video. Currently none of this is secure.
The web based viewer will be easier to create this in. The downloadable viewer is still
basically in the stone age here. Chat and IM work generally well. Voice is a real
problem, in not being dependable and having license issues. We need to develop a new
voice system for OpenSim that works well and is Open Source.
J. Able to have the server functions able to be spread among several servers with
redundancy amongst them. MOSES is somewhat set up to work this way now but more
work needs to be done on this. HighFidelity spreads the server duties out amongst the
clients making the system as a whole much faster. The load balancing is done
considering the ability of the client. Things like cell phones would not take on very much
server duties as a big desktop with a large video card.
K. Create a system within the OpenSim server that offloads much of the work onto the
client. The way it works now, a lot of the grunt work is borne by the server handling
clients. If the load was spread out among clients by ability, the regions will be able to
handle much more avatars before the frame rates go down to unusable levels. The
server needs to be more of a traffic cop who directs info where it needs to go and
handles logins and that type of thing. (This needs further thought to work correctly!)
L. We need to be able to create a good easy way to import objects from Unity and
Maya, MakeHuman and other sources.
M. Security needs to be end to end and heavy duty. A web viewer can use SSL and the
database can be encrypted. At this point the system just has a million holes. Another
reason for updates on outdated code parts! all the security holes need to be plugged
using good Open Source methods.
N. OpenSim needs to be brought up to Enterprise Grade. It seems that in the past
OpenSim was more of a toy for a few on the inside. OpenSim has evolved despite those
attitudes, but now is the time to make it able to handle Enterprise level use and come
out of the perpetual Alpha-Beta cycle. We need version 1!

3. People and attitude Issues:

A. The dirty little secret about OpenSim has always been that there are more than one
version of it out there for each release. One that works well and lesser B grade one that
has purposely thrown in errors to hobble it in some ways or make it inferior to the A
grade release. The A release is used by certain people in the inner circle and everyone
else gets the B grade release. One of the things that I like about MOSES is that there
appears to be only one release that is shared with all. This is what OpenSim needs to
be a success. Purposefully degrading some releases will never allow it to be an
enterprise grade product.
B. OpenSim needs to be treated in a serious way by all who are working on it. It started
out as a reverse engineered version of Second Life based off a viewer that they
released to Open Source. OpenSim seems to be more of a personal toy to mess with
rather than a serious project. There needs to be a strong guiding body, maybe like
Mozilla or the Apache Project has and a group of people seeking funding and grants to
pay for development and refinement. Up to this point development, except the MOSES
project, has been by dedicated coders who volunteered their time to create new
advances in OpenSim. We need to have a small group of coders working on this being
paid for their work. We also need to get this into academia through having interns or
coding class students working on it for credit. If they are good then they can maybe
work for the foundation for pay on future designs.
C. Evangelization and getting people into Virtual Worlds. Most people, including people
in the coding community have never heard about Virtual Worlds, or of OpenSim. This
needs to be changed. In high schools and colleges there needs to be classes made to
introduce students to the potential that Virtual Worlds offer and get them excited about
it. The Raspberry Pi Foundation has done that very well with their little computer. We
might even train coders with OpenSim on a Raspberry Pi! In some ways this is the most
critical thing we need to do.
D. We need to get Second Life out of our life! OpenSim has been following SL since the
beginning. We do have a few things that we have and they don’t, like IAR and OAR files
and HyperGrid. Being a follower does no good. We need to innovate on our own. We
need to put our dreams and great ideas into OpenSim and let it fly. OpenSim is around
10 years old now and in some ways is old technology. There are tons of content out
there that people have made for OpenSim. What needs to happen is develop OpenSim
keeping an eye to compatibility with the future. The viewers don’t need to work in SL!
E. OpenSim needs to be totally Open Source with no DLL or license issues. Everything
in it in it’s base form needs to have an easy to understand and very open license. There
should not be any questions about who owns what or what the license actually is. Any
DLL’s that are maybe someone else’s need to be rewritten or the legalities worked out.

4. Avatar issues

A. Avatar appearance is really quite good at this stage. (at least to the casual observer!)
In order for the system to be more widely accepted there needs to be an easy way of
replicating oneself as an avatar using a mesh system. This would eliminate the seam
issues we have now.
B. Avatar movement at this point is pretty bad and really clunky. The hands and fingers
are basically one piece that really don’t move well and the fingers do not move
separately. In order to move at all one must wear a HUD (A pile of scripts and an
interface you click on to control movement). The HUD is hard to use and not at all
intuitive. The avatar needs to move fluidly like a real person and also have intuitive
movements like a real person does. The fingers need to be able to grip and make a fist
or a V for victory. This is one area where a machine intelligence could really improve
how this works. It should sense when you are going to sit down and just do it! This could
be partly controlled by keeping track of where you are looking. You also need to be able
to shake another avatar’s hand and hold hands and put your arm around them.
C. Avatar dressing at this stage is sometimes a frustrating waste of time spent in
adjusting and then adjusting again. Clothes and accessories need to just go on and be
in the right place. People that are new to this tech need to look good quick! One simple
way would be to have 100 or so good looking male and female avatars pre-built and
available for them to choose from. When you are first going into a virtual world you can
check M or F and and you can get a random avatar to start out with. You can then
change it later if you want to.

5. OpenSim needs to be monetizable!

A. When something is Open Source you have to be a bit more creative to be able to
make money from it. The reason that many Open Source projects fail is that the people
who are working on them just get sick of working for free, or they just have no time
because they finally got a real job! In order to make OpenSim what it needs to be we
need to work on getting funding to pay for the coding that needs to be done. We also
need to get students involved in building modules and add-ons as class projects.
Getting students involves actually helps in more than one way. First is that is a good
way to learn more advanced coding and it is a really good way to get them into Virtual
Worlds and all the really extensive uses there are for them. The biggest problem right
now is that nobody out there has ever even heard of OpenSim or even SL or Virtual
B. At A Dimension Beyond, Inc. we are working on creative ways to be able to make
money from our virtual creations and OpenSim. We realize that there will be no income
from our labor for a while until we can get funding and technology advances a bit more.
We could go the Red Hat route and offer tech support and installation of our systems
and make money from customization. What we really do need to do is think out of the
box a bit. That is what we are working on right now. One good idea is work with some
companies on a trial program that they can do for very little outlay that will show them
just what we can do for them. Because nobody knows what we have they have no idea
of what it can do!
C. There are a few places that use Virtual Worlds for education. Why cannot we get the
Computer Science students into it in the classroom? Running OpenSim on a small
computer like Raspberry Pi would be a good starting place for this. The Raspberry Pi
Foundation in the UK has a ton of educational programs that even little kids can use to
learn coding. Why not set up an image file for Raspberry Pi that would give them
OpenSim pre installed? We should also have free classes in server setup and system
design and building and such. These could also be integrated into the curriculum down
to even the grade school level.

6. OpenSim and the handicapped.

A. OpenSim as it is offers some level of handicap services but very little. Our aim should
be to have someone like Stephen Hawking be able to use it easily. Having it work for
people of different abilities should be a top priority in our design of OpenSim. This might
be a way of monetizing our system. I have discovered that there are many companies
out there that have developed many really good systems that we could partner with and
integrate their tech into OpenSim. One good example is a company in Phoenix that has
a system that follows where you are looking to move the mouse and use an onscreen
keyboard. This is an incredible thing as to use this and be inworld a person would only
have to use their eyes. If nothing else worked this would enable them to have a good
virtual life where they could walk, run, dance and fly.
Another idea involves machine learning and intelligence. There is a company that
has developed a system that can read how you are feeling by your facial expressions.
Couple this with an intelligent NPC and you have a system that could be a best friend
and mentor for you in-world. The list of things you can do for people with different needs
is extensive. You only need to think out of the box!
A note on this: Not every idea in here is mine. A lot of them are from good friends like
Myron, Agda, and Scott who I have worked with extensively. There are many more also
who have helped along the way!

News and Notes


    The Hypergrid WIP Show

    The Hypergrid WIP is a one hour "show & tell" of works in progress or recently completed. All builders from beginner to pro are invited.  Presentations are in voice and text.  For text presentations, best bring the text in a notecard and paste it into chat.  Voice presentations may be captured in video.  Stills and videos from the show may appear in this blog and elsewhere.

    Next WIP show 

    • Next WIP show: Sun. Sept 25, Noon SLT 
    • Cookie II location (fourth Sunday of the month)
    • HG address below: paste into the World Map next to Find. Click Find, TP
    • II 
    • in Kitely: paste into Nav (top) bar of Firestorm, Enter.
    • hop:// II/68/369/22
    • Narasnook  location (second Sunday of the month)
    • Pandora allows presenters to run high threat OSSL functions.
    • Put the line above in your World Map next to Find.  Click FindTP
    • At Narasnook, use World Map to search for Pandora
    • Cookie II location (fourth Sunday of the month)
    • HG address below: paste into the World Map next to Find. Click Find, TP
    • II 
    • in Kitely: paste into Nav (top) bar of Firestorm, Enter.
    • hop:// II/68/369/22

      Previous Articles from the WIP show 

      HG links-- depending on your interests 

      Radio in the virtual world

      Metaverse beginner help

      Schools in virtual worlds

      • I publish 5 or 6 days a week, skipping 1 or 2 days in midweek
      • For more on topics like this, follow Selby Evans in Google+ 
      • What do we do in Virtual Worlds? 
      • Google search this blog: Search bar, upper left--or:
      • Put at the end of the search terms 
      • Annotated screen shots made with Jing
      • All original content on this blog is Creative Commons License, attribution only. 
      • Second LifeLindenSLurl, and SL are trademarks of Linden Research Inc. 
      • This blog is not affiliated with Second Life or anything else. 
      • Ads are from Google.

      No comments:

      Post a Comment

      Note: Only a member of this blog may post a comment.