[Opensim-dev] future rexviewer merger

Ryan McDougall sempuki1 at gmail.com
Sat Dec 6 10:18:28 UTC 2008 

There are a great many comments made while I was asleep, so I hope
you'll forgive me if I try to address them all in one place.

On Sat, Dec 6, 2008 at 3:47 AM, Melanie <melanie at t-data.com> wrote:
> Hi,
>
> I am not opposed to Rex technology. I'm not opposed to have it in
> OpenSim.
> I'm also not opposed to the people who I know (like you).

Hehe, you make me blush. :)

> What I am opposed to is the strategy of doing Windows-only
> development behind closed doors, without community (OpenSim dev)
> feedback, then presenting huge patches that are so brittle that they
> need to be applied without scrutiny or go stale.

On that point we both agree.

As I said, that was historical due to:

1. conscious decision to prioritize the largest userbase
2. lack of linux programmers
3. lack of experience with running a community

We have had an internal reorganization where our steering group
re-prioritized open source. I have now re-prioritized cross-platform,
and added linux programming experience (however we still lack mac
programmers, this will just have to be fixed by the community).

> I would welcome Rex involvement if the Rex developers would join us
> in -dev, discuss things, and submit small patches that we can
> analyze and become familiar with, and modify so they don't conflict
> with the directions where OpenSim core wants to go.

Yes. I am here representing reX as the principal architect for reX
precisely to solve this problem. Our new priority on community and
code quality means more documentation and discussion.

There may be some points where reX and OpenSim disagree on direction,
but thats nothing new, and we can make our changes

>
> Fromn what I gather, Rex is intent on pushing both the login service
> and the viewer onto OpenSim. Including going to the point of
> breaking SL viewer compatibility. I am totally opposed to this. I
> would never endorse a single LL-Code based viewer as _the_ viewer,

Adam addressed this. I don't see that as being a reX goal at all.

> to the exclusion of any other, unless it's fully BSD licensed, e.g.
> developed from scratch.

That's another discussion. :)

> Also, I see anumber of abuse scenarios possible with the
> login/avatar service (commercial abuse), which is why I personally
> favor a different model. Specifically, to keep the avatar appearance
> data on the client and not on any server at all.

Lets keep talking about this, but perhaps on a different thread.

> That is what I can write out right now, there is a diffuse feeling
> of opposition in me, that I have yet to fathom.

Well lets create a diffuse feeling of love. OHhmmmmmmmmmmmmmmm....

> If you take it in small steps, I will be with you. Megapatches would
> be an issue for me.

Agreed.

> Melanie
>

Cheers,

>
> Ryan McDougall wrote:
>> On Fri, Dec 5, 2008 at 10:35 PM, Melanie <melanie at t-data.com> wrote:
>>> In my opinion, nothing. Nothing, that is, that we could not get
>>> without ReX involvement, and better.
>>>
>>> I would like to see OpenSim do an OpenSim solution to the mesh and
>>> asset/inventory system. I don't feel comfortable with the ReX
>>> solution, and even less comfortable with the parts of it they may
>>> hold close to the chest and spring on us later. I expect some
>>> unpalatable bits there. I would rather see us grow to that level in
>>> a true collaborative Open Source, Free environment, without being
>>> under the control of a single corporation.
>>
>> Which corporation is that? Do you know anything about legal or funding
>> structure of reX? Do you know any of us personally?
>>
>> Its a lot to presume with little in the way of fact.
>>
>> How about instead of playing games you just ask me and I'll tell you.
>> If I had the faintest idea what your concern is I'd pre-emptively tell
>> you...
>>
>>> Melanie
>>
>> Cheers,
>>
>>>
>>> Diva Canto wrote:
>>>> As Rex's extensions move closer to opensim (I'm in the group of people
>>>> who can't wait to see meshes and better graphics!), it will be good to
>>>> know more of the details of Rex. I'm not sure I understand completely
>>>> the use of "avatar" here, as we know that opensim uses it to denote a
>>>> pixelated carcass, that's neither a user nor an agent. And in Rex the
>>>> "avatar system" seems like it's more of a user storage/asset system.
>>>> Part of the user's storage/assets includes the pixelated carcass for
>>>> opensim-based worlds that care about that; but it includes all other
>>>> assets owned by the user.
>>>>
>>>> It would be interesting to compare existing efforts in opensim such as
>>>> the Hypergrid and the distributed asset server with Rex's avatar system.
>>>> Is it possible that
>>>> Hypergrid+DAS >= Rex's Avatar System ?
>>>> What else does the avatar system bring to the table?
>>>>
>>>> Crista
>>>>
>>>> Ryan McDougall wrote:
>>>>> Thank you for your addition to the discussion James, however I think
>>>>> Paul may have misunderstood reX architecture when made a private
>>>>> discussion public without consent.
>>>>>
>>>>> The reX avatar and authentication servers replace the User server and
>>>>> split it into Avatar storage for reX's own avatar system, and pure
>>>>> authentication of identity (the latter a bit like OpenID). The idea is
>>>>> to enable avatar portability: the ability to take the same
>>>>> meticulously created reX avatar from one grid to another.
>>>>>
>>>>> Afaik there is no analogue for stock OpenSim, and thus its difficult
>>>>> to merge until the reX team is better integrated with OpenSim.
>>>>>
>>>>> Moreover I see room for improvement there, and have some ideas to
>>>>> change it further. It would be better for all parties if we did our
>>>>> prototyping first, and then worry about arguing who has the better
>>>>> concept or code later and the discussion is on objective ground.
>>>>>
>>>>> Cheers,
>>>>>
>>>>> On Fri, Dec 5, 2008 at 8:05 PM, James Stallings II
>>>>> <james.stallings at gmail.com> wrote:
>>>>>
>>>>>> LOL an ESPECIALLY good idea given they are the originators of the project
>>>>>> and maintainers of the core offering :D
>>>>>>
>>>>>> Cheers
>>>>>> James
>>>>>>
>>>>>>
>>>>>> On Fri, Dec 5, 2008 at 11:52 AM, Jani Pirkola <jpirkola at gmail.com> wrote:
>>>>>>
>>>>>>> Paul,
>>>>>>>
>>>>>>> Good idea to involve Opensim-dev, since they might have better ideas. I
>>>>>>> suppose Adam Frisby has already thought how or if to integrate avatar
>>>>>>> storage to Opensim.
>>>>>>>
>>>>>>> One reason for avatar appearance (or agent domain) handling being at the
>>>>>>> same place with the opensim is because in SL this was so. When we move
>>>>>>> towards free 3D Internet (free as libre), then we need to be able to take
>>>>>>> our avatar with us from grid to grid. This means that a grid must accept
>>>>>>> foreign agent domains - and eventually someone would like to run just the
>>>>>>> avatar service.
>>>>>>>
>>>>>>> Best regards,
>>>>>>> Jani
>>>>>>>
>>>>>>> 2008/12/5 Paul Fishwick <metaphorz at gmail.com>
>>>>>>>
>>>>>>>> I am resending this since i think I sent a message from the wrong
>>>>>>>> email account...
>>>>>>>>
>>>>>>>> ..........
>>>>>>>>
>>>>>>>> Jani
>>>>>>>>  I confess to not knowing the details of the rex viewer source, so please
>>>>>>>> take
>>>>>>>> my comments in that light. However, to me, it would seem more logical to
>>>>>>>> evolve the already existing UGAIM+R services of OpenSim rather than
>>>>>>>> inventing a separate way of handling region and user services. I feel
>>>>>>>> that
>>>>>>>> we may need more discussion with the opensim developers (which is
>>>>>>>> why I am cc'ing them).  If I am missing something in the logic of your
>>>>>>>> revised
>>>>>>>> Rex viewer implementation strategy, please let me know, and thank your
>>>>>>>> team again for its excellent viewer capabilities.
>>>>>>>> In summary, if there is something about the way that opensim handles
>>>>>>>> avatars or authentication, then suggest fixes to this via opensim-dev.
>>>>>>>> Yes?
>>>>>>>> If the DB schema(s) need to change, then suggest changes or add new
>>>>>>>> tables.
>>>>>>>> -paul
>>>>>>>>
>>>>>>>> Jani Pirkola wrote:
>>>>>>>>
>>>>>>>>> Exactly, that is what we will do. However some of the services need to
>>>>>>>>> be able to run separately from the region or opensim server. The best
>>>>>>>>> example of this is the avatar storage and user authentication, which we feel
>>>>>>>>> needs to be its own entity.
>>>>>>>>>
>>>>>>>>> Jani
>>>>>>>>>
>>>>>>>>> 2008/12/5 Paul Fishwick <fishwick at cise.ufl.edu
>>>>>>>>> <mailto:fishwick at cise.ufl.edu>>
>>>>>>>>>
>>>>>>>>>     Presumably, the "services" aspect can be worked out as a region
>>>>>>>>> module
>>>>>>>>>     for opensim? Either that, or the rex team would work with the
>>>>>>>>> opensim
>>>>>>>>>     team on the opensim trunk to extend its service capabilities to be
>>>>>>>>>     more
>>>>>>>>>     consistent with rexviewer. That way, developers or users are free
>>>>>>>>>     to update or
>>>>>>>>>     re-build opensim whenever it suits them, and rexviewer would still
>>>>>>>>>     function.
>>>>>>>>>     -p
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>     Antti Ilomäki wrote:
>>>>>>>>>
>>>>>>>>>         Actually that's not exactly the case, viewer modifications are
>>>>>>>>>         important to us, but the avatar portability service (and
>>>>>>>>>         participating
>>>>>>>>>         in world server development as well) is another central feature
>>>>>>>>> we
>>>>>>>>>         have. Our system currently allows full avatar portability
>>>>>>>>> between
>>>>>>>>>         different worlds and even the inventory works, at least to an
>>>>>>>>>         extent.
>>>>>>>>>
>>>>>>>>>         2008/12/5 Paul Fishwick <fishwick at cise.ufl.edu
>>>>>>>>>         <mailto:fishwick at cise.ufl.edu>>:
>>>>>>>>>
>>>>>>>>>             I could be wrong, but i thought the point was to make Rex
>>>>>>>>>             purely
>>>>>>>>>             a visual client (viewer) and to let OpenSim handle the
>>>>>>>>>             services?
>>>>>>>>>             -paul
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>     --     Dr. Paul A. Fishwick           E-Mail: fishwick at cise.ufl.edu
>>>>>>>>>     <mailto:fishwick at cise.ufl.edu>
>>>>>>>>>     Dept. of Computer & Info       Phone & FAX: (352) 392-1414
>>>>>>>>>     Science and Engineering       WWW:
>>>>>>>>>     http://www.cise.ufl.edu/~fishwick
>>>>>>>>>     <http://www.cise.ufl.edu/%7Efishwick>
>>>>>>>>>     University of Florida          (PGP Key available at above WWW
>>>>>>>>>     address)
>>>>>>>>>     P. O. Box 116120
>>>>>>>>>     332 Bldg. CSE, Gainesville, FL 32611-6120
>>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> Opensim-dev mailing list
>>>>>>>> Opensim-dev at lists.berlios.de
>>>>>>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Opensim-dev mailing list
>>>>>>> Opensim-dev at lists.berlios.de
>>>>>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> --
>>>>>> ===================================
>>>>>> The wind
>>>>>> scours the earth for prayers
>>>>>> The night obscures them
>>>>>>
>>>>>> http://osgrid.org
>>>>>> http://del.icio.us/SPQR
>>>>>> http://twitter.com/jstallings2
>>>>>> http://www.linkedin.com/pub/5/770/a49
>>>>>>
>>>>>> _______________________________________________
>>>>>> Opensim-dev mailing list
>>>>>> Opensim-dev at lists.berlios.de
>>>>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>>>>
>>>>>>
>>>>>>
>>>>> _______________________________________________
>>>>> Opensim-dev mailing list
>>>>> Opensim-dev at lists.berlios.de
>>>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------------------
>>>>
>>>> _______________________________________________
>>>> Opensim-dev mailing list
>>>> Opensim-dev at lists.berlios.de
>>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>
>>> _______________________________________________
>>> Opensim-dev mailing list
>>> Opensim-dev at lists.berlios.de
>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>
>> _______________________________________________
>> Opensim-dev mailing list
>> Opensim-dev at lists.berlios.de
>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>
>>
>
> _______________________________________________
> Opensim-dev mailing list
> Opensim-dev at lists.berlios.de
> https://lists.berlios.de/mailman/listinfo/opensim-dev
>

More information about the Opensim-dev mailing list