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

[thai low (endivatomic.eu)]

I'm new to this board and relative new on OS too, so i don't know if i'm 
allowed
to explain my point of view here. If i'm not please forgive me and trash 
this mail.

I spent last year and half working by myself in different direction 
(translating code
and playing with web browser as viewer), waiting the days i would have 
something
concrete to show before talk with somebody, but reading this board i 
decided myself
to say some words too.

In first instance i agree/disagree with Dahlia (3.66??). I agree <<user 
community
wants and expects this compatibility above all else>>, but they are 
users (and me too!),
it mean we like to use software in the way we are use to, not that we 
care about how
this goal it's reached (and this is true for OS as for any software).
I don't think <<we are probably nearing crossroads>>, i think we are in 
delay and if we
don't make something "energic" all the job made to OS will be slowly 
lost, it will remain
a nice game or a <<nice piece of software>> until a new platform will born.

I agree with Doug: an organization it's a need, to prioritize goals in a 
long term view,
able to collect requests and ideas and make a filter about what could 
came first and what
later. I think Terry was really active in this direction when he created 
Opensimunited.

Back in concrete, in Doug's style, i'll throw what the goals i would 
like to follow:

1) Complete rewrite of Robust in different language (maybe php like 
Simiam) to separate
    "management" from "operativity", powering the development of 3rd 
party web based
    interfaces. All with <<Clear and up-to-date API documentation>>.

2) slow translation of simulator into c++ for better performance and 
integration with
    other engines (i must tell that i'm not a c++ coder, i'm not talking 
about this for
    a kind of mania).

3) Start definitions of a client development, a sort of proxy to install 
on single clients
    making bridge between Web Browsers and simulators OR set the basis 
for a new viewer.
    I've nothing to say against actual viewers, they are wonderful but 
they must follow SL
    directive in first. Without having a new base code for a new viewer 
it would be
    impossible to get to point 5.

4) Clear and up-to-date API documentation.
    Without this, an open software can't be called "open". Like Cinder, 
i felt exactly the
    same frustration losing more time in trying to understand how to 
dialogue with OS
    then writing code i wanted.

5) Do not forget compatibility with.... but forget SL. This is 
OpenSimulator not a
    brand of Lindens Labs.

This are just 5 points as pure example, i don't go on just because i 
think i bored all
of you enough.

All this to say that, if a new organization will born, well structured 
and strongly motivated,
i would be really happy to give my contribution in term of code lines.
To don't make names... Terry, why don't you organize something in this 
direction?
... it could also be a fork at beginning, to have a check if it will 
work and go on with its
legs without disturb the main Opensimulator project.

Marco Montrasio
inworld: thai low