[Opensim-dev] The Future of Open Simulator(?) (UNCLASSIFIED)

Tue Aug 11 20:44:48 UTC 2015

Thank you Doug for sharing your insights. I do tend to agree to some extent
that we are probably nearing crossroads and I'll try to explain why.

I've been involved in this project for around 8 years and I had been using
and contributing to libOpenmetaverse for some time before that. I've been
an avid user and code contributor and core developer. I've spent much time
testing, supporting users, and recruiting and supporting developers. I've
worked on many related projects that are not part of the core code base.
I've participated in administration and planning. As such I've developed a
intuitive sense of the implicit roadmap which, for lack of a better
description, is to emulate Second Life as closely as possible and to add
new features which LL may not wish to develop, all while maintaining the
best compatibility possible. This deviates from the project description on
http://opensimulator.org but I believe the user community wants and expects
this compatibility above all else. This has been demonstrated to me
countless times via my interactions with users and one only needs to peruse
our Mantis system for a few hours to gain a similar understanding. Those
who depend on this compatibility are likely not vocal about it until
something breaks it. at which point they tend to complain en masse.

I see the crossroads ahead due to LL's seeming lack of future development
for Second Life. OpenSimulator should continue but there are some
difficulties which could impede progress towards the beyond: the user need
for compatibility and the architecture of the core code which is very
specifically designed around the SL way of doing things. I also see a lot
of interesting ideas expressed by the community; some which come to mind
are integration of voxels (anything from simple object editors to a
complete voxel-based world), space based simulations with planets and
altered physics, and augmented and/or virtual reality where real-life
content can be mixed with virtual and which can provide a very immersive
experience when used with such technology as a Oculus Rift or others, to
name a few. Such enhancements are likely huge, long term projects on their
own and may turn out to be so fundamentally incompatible with the current
code and each other that all becomes impossible. Or perhaps some clever
developers can invent ways to make it all work together. Developing a
feasible architecture and a roadmap around such features would likely not
be a simple exercise.

There are other goals shared by some users based on improving reliability,
scalability, performance and operating cost, I tend to consider the MOSES
contributions to be along these lines. I believe such are more feasible but
probably less likely to prevent our reaching the crossroads I've eluded to
above. They are still valuable goals and their persuit will likely bring
benefit to most users.

I tend to agree that losing Overte is undesirable. Unfortunately I lack the
time and resources to contribute towards maintaining it. There are
alternatives that have been discussed and one of them may eventually come
to be. I really can't offer any predictions in this regard.

Please note that I am offering my own opinions here and I am not speaking
for other core developers or contributors. We're a loose bunch and we don't
often speak for each other. I'm sure others may disagree with my statements
but that's OK, such disagreements have helped to make OpenSimulator what it
is today :)

On Tue, Aug 11, 2015 at 9:59 AM, Maxwell, Douglas CIV USARMY ARL (US) <
douglas.maxwell3.civ at mail.mil> wrote:

> Classification: UNCLASSIFIED
> Caveats: NONE
>
> Projects evolve.
>
> I couldn't begin to estimate the amount of work that has gone into this
> valuable project.  The potential for technical and economic success is
> profound and I see a bright future for the Open Simulator.  That said, I
> fear
> we are at a crossroads at this time with this project.
>
> It is unclear at this time what the maintainers of the Open Simulator code
> have planned for the project.  Is there a roadmap or some sort of
> goals/objectives you are working against?  What development targets would
> you
> like to see met in 12, 16, and 24 months from now?
>
> The MOSES project has needs & requirements that we are stepping up and
> supporting with internal development, but we aren't the drivers for the
> Open
> Simulator project.  We've done our own internal gap analysis and determined
> where in the OS code there should be investment in stability, monitoring,
> and
> scalability improvements.  In short, we are returning our code to you to
> adhere and abide by applicable derivative source code licensing terms.
>
> I believe the removal of the Overte as a formal governing entity is a
> mistake
> if you plan to encourage participation from business and government.  The
> CLA
> was viewed by my organization as a formalized relationship acknowledging
> the
> legal responsibility of open source code stewardship and use.
>
> If this were simply a hobby, then Overte and the CLA would not be needed.
> However, the Open Simulator is being used by businesses charging money for
> service, by researchers studying human behavior and technical behavior, by
> educators, and more.  Like it or not, you have created a product that needs
> management and attention at a higher level than the ad-hoc method that is
> currently your standard operating procedures.
>
> Project management must evolve.
>
> As projects are started at the grass roots and then emerge as valued
> commodities, the need for different styles of management is required.  A
> project with two active developers is different than a project with 20 or
> 200.
> If the management does not evolve, then the project will be limited and
> growth
> is not possible.  I encourage you to think about a new structure that can
> handle influx of large amounts of donated code in a short time.  The kinds
> of
> investments needed to make this a world class simulator requires you to
> step
> up and begin project planning.
>
> This is a community effort.
>
> If the community values this work and would like to see it grow or even
> receive maintenance, then the community must voice.  This code does not
> belong
> in the hands of a gov't agency or corporate entity.  This code belongs in
> the
> hands of a strong non-profit that can handle grant and contract funds to
> pay a
> staff of maintainers, code reviewers, testers, and functional area code
> managers.  This could be an Overte spin-off, or even an academic
> institution
> of some kind.
>
> I've given you a glimpse into what the next 9 months of development for the
> MOSES related Open Simulator issues.  We came in this spring at a time when
> development seemed to be winding down and things were quiet after the 0.8.x
> releases.  What will you do when we reach the logical conclusion of our
> work?
> What is next for Open Simulator?
>
> I look forward to your feedback and constructive discourse.
>
> v/r -doug
>
> Dr. Douglas Maxwell
> Science and Technology Manager
> Virtual World Strategic Applications
> U.S. Army Research Lab
> Simulation & Training Technology Center (STTC)
> (c) (407) 242-0209
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> _______________________________________________
> Opensim-dev mailing list
> Opensim-dev at opensimulator.org
> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://opensimulator.org/pipermail/opensim-dev/attachments/20150811/43c69f42/attachment.html>