Open Source Second Life – The Geeks Strike Back

OS Grid LogoRecently, Gareth Ellison, one of the promoters of the Open Source Grid, did a promotional event by being in-world to explain how exactly the OS Grid works and how people can connect to it.

The OS Grid is one of several that run the reverse-engineered (but Linden-approved) Open Grid Services / OpenSim software suite. Each has a different goal, but OS Grid is strongly encouraging the creation of a completely free environment, and want to establish a foundation to draw resources (financial and human) to support a free, costless, Second Life-compatible grid.

You can connect to the OS Grid with any standard SL client, or compile your own SL-compatible client. The beauty of it is that it relies on the almost-completely-documented SL Protocol. In effect, these SL-compatible grids are very similar to what LL is doing with their own grid: server and client are pretty much independent, but both agree on a common protocol, which is going to remain stable for a while. So LL can launch new clientes without changing the server software; and the reverse, of course, is also true.

So how advanced are these reverse-engineered solutions? Well, the mere fact that they work at all is a pleasant surprise. OpenSim allows you to run your own server, using applications running on top of libSecondLife; Open Grid Services is the “glue” that replaces LL’s “central servers” (login and asset servers, etc.) and make several disjointed sims be part of the same grid. When you register with OS Grid, you can pick any avatar name and join any sim that is on the grid. Similarly, if you managed to compile and run a version of OpenSim on your computer, you can add your own sim to the existing grid (note that avatars will require around 100 Kbps of upstreaming bandwidth, so forget about running your own sim behind an ADSL/cable modem connection — it’ll handle 3-4 avatars at most).

Right now, obviously, don’t expect “miracles”. OS Grid allows you to have an avatar and talk to other people, and even IM them, but there is not much more to do beyond watching the landscape — which will also be free of prims, since OS Grid hasn’t implemented an asset server yet.

Also, the first question that everybody asks is “will I be able to connect my sim to Linden Lab’s grid?”. The answer, of course, is no. Linden Lab does not allow anyone to connect their own sims to their grid — something that will possibly change in 2009/2010, but not before — so basically this is just one separate grid which runs the same protocol as Second Life, but there is no way to get your inventory from LL’s servers and upload it to the OS Grid. At least — not yet!

So why would you use it? Right now, it’s very likely that you won’t; OS Grid is about at the same level of development as the LL grid was in, say, 2001 or so. But it has a huge advantage: more programmers than LL had back in 2001; a fully-developed open source client (so people can focus on the servers for now); and, of course, we all know what the final result is going to be! LL, back in 1999-2001, had no clear idea on how the grid would work in the future at all.

An open source grid is naturally the dream of everybody who’s tired with LL’s recent strong measures in limiting personal freedoms, which will only increase in time. By distributing grids all over the world, and interconnecting them together, the Metaverse will just become like the Internet: independent of local authorities which can have global reach. Put into other words: you’ll be allowed to do whatever is legal for your country and the country where you’re connected to (like on the Internet), but nobody outside your country will be able to enforce anything you like. And, of course, if your country is restricting your personal freedom too much, you can jump over to the sims hosted on a different country. Just like on the Web.

It also means that even if Linden Lab is shut down due to legal and governmental pressure to turn Second Life into Disneyland, we have now a way out: the software is there, and it works. It only needs a lot of work until it becomes more and more similar to LL’s own SL grid, but at least we know what needs to be done. This is the first step towards Second Life’s immortality: making sure that Linden Lab’s eventual downfall will not leave the almost 9 million registered users without a place to go.


I’ve attached the transcript of the meeting below:

Gareth Ellison: extropia: i wish the WTH (World Transhumanist Association) wre more proactive

Laetizia Coronet: well I had better run off before the new talk starts

Gareth Ellison: they have the philosophy down

Books Janus: Gareth Ellison, you have me so baffled. You’re the ball.

Gareth Ellison: they need the practical technology down

Gareth Ellison: that orange ball next to me is an orange btw 😉

Books Janus: lol

Tracer Ping: to make snow crash into a game you have to have interesting people

Morgaine Dinova: Well, transhumanism is pretty much on hold, waiting for nanotech, which got delayed by Bush and vested interests. Annoying

Laetizia Coronet: you see? WHen I leave the light goes out 😛

Gareth Ellison: nanotech isn’t needed

Extropia DaSilva: Anyway, Extropia means ‘a dynamic environment where people choose the social form they prefer’ which could be a description of SL/metaverse.

Gareth Ellison: i hacked up a simple BCI myself at home

Gwyneth Llewelyn: hehe

Gareth Ellison: and just took a big dose of piracetam earlier

Morgaine Dinova: Gareth: it’s not needed for handwaving, but it’s needed for engineering.

Gareth Ellison: lots of things can be done without nanotech

Morgaine Dinova: Not enough

Lalinda Lovell: like hoovering

Tracer Ping: all the tech in the world can’t stave of the boring consumer march for entertainment

Gareth Ellison: when i have the cash my hackish BCI will replace my keyboard

Laetizia Coronet: bye everyone – next time I’ll try to make an untrustworthy avatar

Books Janus: bye, laetizia

Books Janus: don’t wear a tux.

Extropia DaSilva: Technically-speaking the construction of the matrioska brain does not need nanotech. Just that it would not be optimal without it.

Morgaine Dinova: Laetizia: hahaha, maybe we should all do that 😛 See you next time

Laetizia Coronet: lol Books

Gareth Ellison: should i just dive into the OSGrid talk if your thinkers group is more or less done?

Gwyneth Llewelyn: hehe Laetizia! but you need to come under a new alt

Lalinda Lovell: was this place on tv?

Laetizia Coronet: bye Morg

Gareth Ellison: only 5 mins to go

Extropia DaSilva: yes that is fine with me Gareth,

Gareth Ellison: ok

Extropia DaSilva: means I get to hear a bit about it.

Gwyneth Llewelyn: well, Gareth, perhaps you should wait a tiny bit, since I believe more people will come just for that event

hoppa Balut: Wells as humanity we decided to chosse for the techway out! So finish that course

Morgaine Dinova: Oh, OSgrid! Cool, I’m staying for that. I thought it was BDSM next …. 🙂

Gareth Ellison: i’ll wait 5 mins then

Lalinda Lovell: charismo are you a character from naked lunch?

Gwyneth Llewelyn: Morgaine, pfft 😀

Scope Cleaver: I’ll stay either way Morg

Gareth Ellison: no, the kinky uploading takes place later – who wants to upload into a dildo?

Joaz Janus: Thx for stimulating chat folks…..I trust you all…no honestly I reaaly do….(she lied)

Morgaine Dinova: You have a quorum already, Gareth

Lalinda Lovell: was this place on tv

Tracer Ping: I jsut want to say chaawkklit rainnnn

Gwyneth Llewelyn: haha Joaz. “Everything I say is a lie”

Gwyneth Llewelyn: 😀

Lalinda Lovell: is this thing on

Gareth Ellison: ok, i’ll start

Joaz Janus: fades to….black

Gareth Ellison: i was asked to come here and speak about OSGrid – the attempt at a totally free and open source grid

Gareth Ellison: running opensim+OGS (Open Grid Services

Gareth Ellison: shall i presume all present are familiar with opensim?

Lem Skall: sorry, brb

Morgaine Dinova: See you Jozie!

Morgaine Dinova: Yup

hoppa Balut: Please explain opensim?

Gareth Ellison: well, for those unfamiliar – opensim is as it sounds – an open source sim

Gareth Ellison: produced by reverse-engineering the SL protocols

Gareth Ellison: much of the work on the protocol level was performed by the libsl team

Gareth Ellison: the sim component was initially produced as a proof of concept

Gareth Ellison: but has now been expanded into a working application

Freida Barzane: So, Sl where you create your own sub-world part of the virtual universe?

Gareth Ellison: alongside the sim software, there is a backend of the grid services

Nix Sands is Offline

Gareth Ellison: known as OGS (Open Grid Services)

Gareth Ellison: that’s the servers which store stuff like user profiles, inventory, assets etc

Gareth Ellison: am i being followed here?

Books Janus: yes

Morgaine Dinova: Yup

Extropia DaSilva: yes.

Flyingroc Chung: yup

Freida Barzane: think so

hoppa Balut: hope

Math Gazov: what do you need to run your own server?

Flyingroc Chung: so, grid services have to be centralized?

Tracer Ping: So this will be passworded and secure?

Gareth Ellison: ok, OSGrid is essentially an installation of the grid services

Gareth Ellison: to which anyone can link their own sim

Gareth Ellison: the idea is that you download opensim and host it on your own server then link it to OSGrid

Sean18 McCarey: and it can be anywhere on the internet

Tracer Ping: A sim without $L sounds like a good idea to me

Math Gazov: how do you run your sim?

Morgaine Dinova: Adam said earlier that the services are per site, so not centralized

Freida Barzane: No land shortage, no real use, at least at first, for standardized currency

Gareth Ellison: within one grid the services are centralised

Flyingroc Chung: I see

hoppa Balut: Sounds like P2P serving

Gareth Ellison: what OGS allows is for anyone to run a grid

Gareth Ellison: and what OSGrid allows is for anyone to run a sim without needing to also run a grid

Sean18 McCarey: neat

Extropia DaSilva: But ‘the grid’ is, what. 10,000 PCS?

Gareth Ellison: anyone who hasn’t already done so i advise to go to the website and get an account

Afn Bade: I think that commerce will be in micropayments not in L$ on open servers that support money transactions

Gareth Ellison: www.osgrid.org

Sean18 McCarey: and grids and sims in other locations can all interconnect?

Math Gazov: will OSG grids connect with SL?

Gareth Ellison: Extropia: the grid is the backend services and the sims

Tracer Ping: database servers

Gareth Ellison: LL have not yet opened their grid to third-pary servers

Extropia DaSilva: How can my primary run SL on her poxy little laptop?

Gwyneth Llewelyn: hehe Math 🙂

Gareth Ellison: *party

Freida Barzane: Would it allow 50 people to meet in an area with no lagfest? Or would it be dependant upon the host’s connection speed?

Gwyneth Llewelyn: in 2009 possibly, yes

Tracer Ping: what kind of strain are we talkign about can someone run this over say a cable connection?

Gwyneth Llewelyn: oh my

Gareth Ellison: it’s dependent upon the connection speed

Flyingroc Chung: sims interconnect with each other in one grid, of which the services are centralized

Gareth Ellison: 38kbps per user

Extropia DaSilva: If LL require tens of thousands of pcs for the grid?

Gwyneth Llewelyn: Tracer, make a quick calculation. One avatar takes, on average, 100 Kbps

Gareth Ellison: extropia: LL have 1000s of sims

Gareth Ellison: they run 4 per server

Gwyneth Llewelyn: if you want to host events with 40 people,

Gwyneth Llewelyn: that’s 4 Mbps — UPLOAD speed

Gareth Ellison: opensim can run 64 per server

Tracer Ping: ok so basically a big money sink

Gareth Ellison: it’s vastly more efficient

Math Gazov: LL is very centralised Extro, smaller servers would hols only your friends maybe

Tracer Ping: to have a private place for friends

Gareth Ellison: different design

Freida Barzane: That is a lot of bandwidth

Extropia DaSilva: Oh OK.

Gwyneth Llewelyn: Aye, it is

Sean18 McCarey: and we can run script the same way?

Books Janus: Does it connect into sl SEARCH

Gwyneth Llewelyn: specially because most ADSL/cable connections in your home will have 128 Kbps-512 Kbps on average

Gareth Ellison: the eventual aim of OGS is to enable various grids to interconnect

Gareth Ellison: OSGrid is one such grid

Freida Barzane: Whoops, just got word time for dinner — take care all

Gareth Ellison: lots of businesses may run their own smaller grids

Math Gazov: P2P, paralell processing?

Gwyneth Llewelyn: Gareth — are you planning to interconnect with the Deep grid as well?

Gareth Ellison: which one may move into

Gareth Ellison: Gwyneth: when foreign user support is done, yes

Afn Bade: deep grid?

Gwyneth Llewelyn: Ok

Gwyneth Llewelyn: Afn — Deep Grid is a “similar” project

Quanta Torok: When you say that LL ‘may possibly open in 2009’ is that from inside sources or ?

Gareth Ellison: deep grid is similar but has mildly different ideaology and goals

Gwyneth Llewelyn: Let’s say, Quanta, that’s from “wishful thinking” of LL’s Board

Gareth Ellison: run by a fellow opensim dev

Tracer Ping: streaming content is just not feasable for the average user

Tracer Ping: If you could make a static database

Gwyneth Llewelyn: aye, Gareth — just wondering if you’d merge projects at some point.

Gareth Ellison: for myself, my goals are to spread the metaverse like the WWW

Math Gazov: SL will have to open soon, if independent grids open and join

Gareth Ellison: not merge – link

Gwyneth Llewelyn *nods* @ Gareth

Gwyneth Llewelyn: Yes… link!

Gareth Ellison: that’s a critical part of the vision

Math Gazov: link!!

Flyingroc Chung: hm, does opengrid also support virtual currency?

Gwyneth Llewelyn: Math — LL hasn’t developed an open source version of the server software yet

Gareth Ellison: did anyone here attend my talk at SL4B?

Extropia DaSilva: This almost sounds like your crowdsourcing the technology of Sl idea, Gwyn.

Sean18 McCarey: has SL stated any official position on such an interconnection?

Math Gazov: didn’t you say 2009 gwyn?

Gareth Ellison: there i cited SL asa precursor

Gwyneth Llewelyn: aye Math

Gareth Ellison: apoloogies for the typos btw – lag

Kaskazi Munro: When distributed sim becomes as large or larger than SL – LL will be “happy” to link.

Gareth Ellison: at SL4B i talked about where i envision the future direction beyond SL

Lem Skall: Gareth, what is the intent then, link with SL eventually, run separately, “replace” SL?

Tracer Ping: your talking fiber optic speeds to end users for this kind of thing

Gareth Ellison: one intermediate step between SL and what could be considered the true 3D web is these open SL-compatible grids

Flyingroc Chung: distributed metaverse

Gwyneth Llewelyn: mm hmm

Alpha Auer: this almost seems to me like the birth of the 3D internet…

Gwyneth Llewelyn watches Morgaine drool

Kaskazi Munro: Bandwidth (I’m told) is growing faster than cpu speed and storage capacity.

Gareth Ellison: remove from centralised control

Gareth Ellison: the aim is to be compatible with SL and extend technical features

Tracer Ping: adobe tried that and failed miserably

Morgaine Dinova: Not droolling until I see the scalability architecture 😛

Gareth Ellison: ultimately, SL will become obsolete

Sean18 McCarey: will there be a standardisation effort?

Ina Centaur: / curious.. if SL were to open source the servers, would they also open source the asset/inventory servers?

Gwyneth Llewelyn: 😀

Gareth Ellison: until then, we need a compatible replacement

Gwyneth Llewelyn: possibly, Ina 😉

Gareth Ellison: LL are likely to open the sims

Gwyneth Llewelyn: very likely, yes

Gareth Ellison: the backend i doubt

Gwyneth Llewelyn: I’d say, the likeliness of them opening the servers grows towards 100% in late 2009 😉

Ina Centaur: / content ownership problems might arise if that occurs

Scope Cleaver: Whats their revenue model then I wonder

Gwyneth Llewelyn: agreed, Gareth. I haven’t heard any talk about the backend servers.

Ina Centaur: / content creators may not want crowdsourced servers having open access to their wares

Tracer Ping: I mean you gotta think, do you really want to reverse engineer such an unstable database server in the first place

Gwyneth Llewelyn: Scope: interconnection agreements

Tracer Ping: or just make something that works

Morgaine Dinova: Ina: you can’t hold back the march of progress just because people’s business models are badly affected.

Gareth Ellison: the protocol itself is not a major problem

Gareth Ellison: the architecture needs updating

Math Gazov: content hosting can’t be a monopoly, that’s the big mistake of SL

Tracer Ping: so it’s hardware related

Gareth Ellison: i have tons of documents detailing how to make everything scalable

Gareth Ellison: it’s not hardware related – it’s software

Kaskazi Munro: How about currency exhange?

Flyingroc Chung: their server software is buggY?

Gareth Ellison: the protocols are stableish, the code is buggy

Books Janus: who’s designing the software?

Morgaine Dinova: On osg.org or opensim.org?

Ina Centaur: @unstable database server… MySQL?

Sean18 McCarey: shouldn’t inventory be on client computers rather than an asset server exclusively?

Gareth Ellison: about L$ – that’s an undecided issue for OGS

Flyingroc Chung: “stableish” doesnt inspire confidence

Tracer Ping: just like any mmorpg that has tons of people workign on the same code

Gwyneth Llewelyn: Ina: MySQL is not “unstable” if properly tuned,

Tracer Ping: bloated ware

Gwyneth Llewelyn: something which is true about almost all software really 🙂

Ina Centaur: @sean, stuff on client computers remain inaccessible if client is off or has low bandwidth

Gareth Ellison: the code for opensim/OGS is being worked on by myself and others

Tracer Ping: but can you remove a feature without ruining it

Books Janus: you update?

Gareth Ellison: if anyone wants to talk to the other devs then get onto efnet #opensim

Morgaine Dinova: efnet? Not freenode? Efnet is ful of kiddies and splits

Gareth Ellison: efnet was chosen for historical reasons

Zany Lane: How many sims are running on open source grids today?

Sean18 McCarey: no reason “assets” can’t be on client too with a system preference for the client thus reducing bandwidth for assets

Gareth Ellison: deepgrid has 300+, OSGrid is quite new and has around 7 at last check

Gareth Ellison: it’s just over 1 week old

Zany Lane: Thats pretty good for just getting started.

Gareth Ellison: deep grid is a few months old now

Gwyneth Llewelyn: Well, to be honest, DeepGrid has perhaps close to 400 *registered* sims,

Gwyneth Llewelyn: availability is however around 1% 😉

Zany Lane: Can anyone log onto them at no charge right now?

Afn Bade: So the goal of OS grid is to replace LL’s sim with a protocol that can then become expanded and multiplatform?

Math Gazov: where’s deepgrid???

Gwyneth Llewelyn: aye, Zany 😉

Sean18 McCarey: i wonder how much open market data centre fees for an opensim server would compare to SL charges

Tracer Ping: good question

Morgaine Dinova: Adam said just a couple of hours ago that Deepgrid was really just his test system, and buyer beware.

Gwyneth Llewelyn: Math:

http://www.deepgrid.com/

Extropia DaSilva: Deep grid was built by pan-dimensional beings in order to compute the answer to THE question…no sorry that was deep Thought..

Gareth Ellison: i could have arranged for this talk to happen on OSGrid actually

Tracer Ping: lets see basic /sl free

Gareth Ellison: OSGrid is guaranteed to remain free

Gareth Ellison: hosting provided by the open metaverse foundation

Gwyneth Llewelyn: yay for that decision, Gareth!!

Scope Cleaver: wow

Gareth Ellison: if i wanted to commercialise it, i’d have to get new hosting

Math Gazov: bandwith, data storing and internet 2 should greatly decrease charges in a couple of years

Gareth Ellison: and i wouldn’t want to anyway

Gareth Ellison: progress > profit

Math Gazov: ty Gwyn 🙂

Flyingroc Chung: how does one connect to OSGrid?

Ina Centaur: viva!

Extropia DaSilva: Gwyn told me Internet 2 is defunct.

Gareth Ellison: a cheap server can be got from some places for $9.99/month

Sean18 McCarey: unlimited bandwidth hosting with unlimited datacentre support?

Morgaine Dinova: Nice to hear that Gareth

Tracer Ping: This isn’t about connecting it’s about creation atm

Ina Centaur: / would the open metaverse foundation have more uptime than LL though o_O

Ina Centaur: @gareth.. dedicated server for $10/month? o_O

Gareth Ellison: if anyone wants to discuss linking a sim and wants technical details/instruction please email me

Gareth Ellison: [email protected]

Gareth Ellison: virtual dedicated

Gareth Ellison: plenty for a sim

Sean18 McCarey: virtual dedicated?

Morgaine Dinova: Xen?

Sean18 McCarey: you can put easily 100 virtuals on a server

Gareth Ellison: run your server inside a virtual machine

Scope Cleaver: Basicaly more than one per physical machine

Gareth Ellison: Xen/UML/whatever

Tracer Ping: thats like 100 12 packs of beer

Sean18 McCarey: dedicated means 1 machine 1 server

Gareth Ellison: it’s trivial to put a few 1000 sims per box

Gareth Ellison: virtual dedicated == dedicated VM

Sean18 McCarey: it makes no sense

Sean18 McCarey: i have never heard that term

Extropia DaSilva: OK I have to go. Bye everyone!

Scope Cleaver: Later Extro *hugs*

Morgaine Dinova: Bye Extie!

Alpha Auer: Bye Extropia

Ina Centaur: / 1000 sims on a box… but at the cost of turning everyone into 1-prim boxes?

Gareth Ellison: the OMF btw is quite new

Flyingroc Chung: how much storage does your current asset server have right now?

Tracer Ping: I don’t know I still think reverse engineerign is the wrong way to approach this

Gareth Ellison: the asset server is currently under development

Tracer Ping: I think using something like a quake3 open source

Lem Skall: Gareth is scalability of servers then based on the number of sims, or the number of logins or what?

Gareth Ellison: Tracer: reverse engineering is the only way to build something SL-compatible

Tracer Ping: with multiple linked worlds as maps

Tracer Ping: oh compatible

Afn Bade: Will it be possible to have OSgrid scale exponentially using p2p or swarm networks? If 1K people or 10k people wanted watch your sim, the system could scale?

Gareth Ellison: scalability on the backend requires some designed but not-yet implemented changes

Ina Centaur: reverse engineering would be assuming that LL won’t release code

Gareth Ellison: once those changes are in, it should theoretically be infiniteelly scalable

Tracer Ping: you really think they will let you run a server that links to the money mill that is the LINDEX?

Tracer Ping: you could create for free and bring in your warez

Gareth Ellison: whether LL ultimately open their grid is of little concern

Scope Cleaver: it is?

Lem Skall: Gareth, I meant scalability per sim, or isn’t a sim even running on a single server?

Sean18 McCarey: also the software license usually prohibits reverse engineering if in fact SL holds any valid copyrights or patents.

Gareth Ellison: the open grids will overcome

Afn Bade: swarm video, audio… you might get 1k viewers… mass market potential there if it can be done.

Lamont Helvetic: if it is interoperable.. the warez will be interoperable anyway

Gareth Ellison: reverse engineering of network protocols is completely legal

Gwyneth Llewelyn: Sean18, actually, LL is pretty much encouraging this…

Gwyneth Llewelyn: Linden employees *help* these guys out 🙂

Tracer Ping: It’s pointless because the databases won’t be of synch

Tracer Ping: you would need a vanilla avatar and db to connect

Sean18 McCarey: not if it copywriten and expressly prohibits it

Gwyneth Llewelyn: A question, Gareth, can I ask you what your thoughts on the implementation of the asset servers are? 🙂

Math Gazov: we need LL connection to let all this content out

Gareth Ellison: you can not copyright protocols

Gareth Ellison: only patent them

Gwyneth Llewelyn rather liked the approach of using DNS, like the Sheep do

Morgaine Dinova: Simple number of sims doesn’t give you scalability. You have to scale for people going to events, otherwise your “scalability” is just an illusion, as in SL currently.

Gareth Ellison: Gwyneth – we have an asset server, but it’s unstable right now

Sean18 McCarey: you can copyright the software and patent the process

Gwyneth Llewelyn: Ah, just curious 🙂

Morgaine Dinova: “Scalable” but everyone having to staty in their home sim is not scalable at all.

Gareth Ellison: software patents are not enforcable in europe, and in the US LL’s patent has been refused

Math Gazov: scale connections is going exponential, that’s what’s messing SL right now

Lem Skall: right Morg, I was thinking even that it’s easy to run 1000 empty sims on a server, but what about one server with 1000 avs on one server?

Tracer Ping: LL is in a world of shit right now

Gareth Ellison: the issue with scalable sims is a flaw of the protocol

Sean18 McCarey: sounds good then

Gareth Ellison: i can’t htink of a way with the SL protocol which could allow distribution of sims

Morgaine Dinova: OMG, the USPTO rejected a patent application? It must be the Second Coming ….

Gareth Ellison: i have privately designed a new protocol

Lamont Helvetic: well.. teh casino closing is a great opportunity to get some donations to the cause

Razor Briggs: sorry i havent been talking just went and signed up at osgrid.org

Lem Skall: what is the architecture? one sim completely on one server or a fixed number of avs per server, no matter what sim they are on? or another possibility?

Gareth Ellison: which may solve these issues, but that’s a long long project

Gareth Ellison: this new protocol i call gmmp

Ina Centaur: +morgaine… yes, what are the conditions where 1000 sims/core is possible?

Tracer Ping: I’m still tryign to figure out what the point is

Math Gazov: reverse engineering allows you more efficient an open protocols

Gareth Ellison: generic metaverse messaging protocol

Tracer Ping: besides bragging rights

Sean18 McCarey: no

Gwyneth Llewelyn: oh nice

Sean18 McCarey: not without massive improvement in server speeds

Flyingroc Chung: what platform are you developing the software on?

Gareth Ellison: gmmp is based on several layers of abstraction – a “channel” may have text, voice, 3D etc

Tracer Ping: I dunno I’m just the guy who is along the lines of why buy a cow when you get the milk for free

Gareth Ellison: and each channel is distributable across the network

Gareth Ellison: opensim is being developed in mono and .NET

Gareth Ellison: in order to make use of libsl

Razor Briggs: could someone IM me who knows about this and can fill me in so i can catch up 🙂

Sean18 McCarey: .net is not open

Sean18 McCarey: it’s commercial

Gareth Ellison: my (dying) C fork uses apache modules

Razor Briggs: but xna’s .net is open

Flyingroc Chung: mono is open

Gwyneth Llewelyn: But Mono is…

Sean18 McCarey: at least SL is all linux

Gareth Ellison: mono is open

Gwyneth Llewelyn: yay 🙂

Flyingroc Chung: you can run mono on linux, I’ve used libsl a little on a mac.

Gwyneth Llewelyn: oh, you were lucky, FR 🙂

Gwyneth Llewelyn: I had no such luck, lol

Morgaine Dinova: Ina: you have to drop the static allocation of resources and use a dynamic allocation structure. LL’s servers just idle when their inhabitants go to events, while the event sims glow red hot and cannot cope. Static allocation just doesn’t work.

Flyingroc Chung: heh

Gareth Ellison: OSGrid is running on a linux box

Gwyneth Llewelyn: well Morgaine… one way to handle it temporarily is to run 100 servers on a single oct-dual-core-CPU hehe

Scope Cleaver: Base allocation on avatar?

Flyingroc Chung: you needed to mondiy their nant thingy, it was a mess, last time I looked at it

Scope Cleaver: 1 CPU per AV!! yay

Razor Briggs: if we use XNA .nets framework we can pull off some of the most advanced graphics and rendering on the market today

Gareth Ellison: dynamic allocation can’t really work iwth the current model of sims allocated to servers

Sean18 McCarey: oh will good luck getting 55000 people to connect to one webserver

Zany Lane: How reliable are the sims currently?

Math Gazov: 1 CPUs avatars could be nearly alive, Extro will be so happy!!!

Sean18 McCarey: each running 40kbs

Gwyneth Llewelyn: 😀

Lamont Helvetic: or grid computing teh way Sun uses the phrase – a network of machines sharing the workload

Ina Centaur: @moragine.. would dynamic allocation of resources across networks lead to data loss (or lag in transfer, thus indirect data loss)

Gwyneth Llewelyn: hmm

Gareth Ellison: for info on how stable is opensim i refer you to

http://bug.openmv.org

Gwyneth Llewelyn: installing OpenSim on Amazon’s S3 infrastructure? :-d

Gareth Ellison: EC3

Ina Centaur: letting amazon take over LL? ;-P

Gareth Ellison: i think that’s it called

Zany Lane: How many avatars have you had in one sim at the same time with opensim?

Gareth Ellison: ugh, lag

Ina Centaur: ec3 is just access to the amazon db, i believe..

Lem Skall: can a lot of the load on the server be moved to the clients? as in P2P?

Gareth Ellison: Zany – in tests, about 30 or 40 works

Gareth Ellison: but those tests are old

Zany Lane: Thats not bad

Ina Centaur: @lem… kinda like the search for ET o_O

Chase Marellan: EC3 si the Elastic Compute Cloud — servers on demand, basically

Morgaine Dinova: Gwyn: temporarily one could use big iron while remaining statically mapped, yes because internal comms on a top end domained Sun say are cheap. But it’s just not viable on cost, and it can’t scale very far anyway, just hundreds, not dozens of thousands of sims.

Chase Marellan: ec2

Zany Lane: Can anyone have their own sim on a home based server?

Flyingroc Chung: Where would an interested developer start, if he wanted to contribute to the project?

Gwyneth Llewelyn: Yes, I know, Morgaine 🙂

Gareth Ellison: look at the bugtracker, fix bugs

Gareth Ellison: Zany – yes

Gareth Ellison: if you have a reasonable connection

Razor Briggs: has there been any changed to sl in these open grids?

Gwyneth Llewelyn: Zany, it’ll be more of a question of how much upstreaming bandwidth you’ve got on your ADSL/cable connection really

Gareth Ellison: the SL viewer is unchanged

Zany Lane: Will it run reliably on a DSL or cable modem?

Gareth Ellison: a few new scripting languages etc are coming

Gwyneth Llewelyn: For one or two avatars, yes, Zany 🙂

Gareth Ellison: should do if you have the bandwidth

Kaskazi Munro: Ruby on Railsim

Morgaine Dinova: The SL viewer HAS to change, naturally, as the sims scale up for large events. Rendering crowds is a research area still though …. but we’ll have to tackle it.

Gareth Ellison: if i can excuse myself here a minute, i need to consume food

Math Gazov: bandwith will replace L& as exchange base

Razor Briggs: one more question will my avie be fresh if i connect to an open sim? or will i have all my stuff?

Gareth Ellison: apologies

Lem Skall: eat the bunny Gareth

Zany Lane: I heard a rumor the the Linux SL client was going to be in Debian or Fedora as an optional package. Is that true, you suppose?

Tracer Ping: just put a paypal donation link on any open source project

Lem Skall: btw, I don’t know how Gareth can hold that bunny and type at the same time

Morgaine Dinova: It’s already in Gentoo, in the secondlife overlay.

Tracer Ping: I’m with bill gates on this one

Math Gazov: me not

Zany Lane: Do you suppose this movement might actually foreshadow the 3D internet?

Gwyneth Llewelyn: Razor: totally blank

Ina Centaur: / “3d internet” — define

Gwyneth Llewelyn: Oh, for sure, Zany 🙂

Gareth Ellison: i return, and do not upset my bunny

Zany Lane: SL might be to netscape as netscape was to gopher

Lamont Helvetic: umm.. it is a requirement, teh 3d internet can’t be the property of a single company

Ina Centaur: / i think when people do serious research, they’d rather browser it than have a bunch of funky av distractions..

Math Gazov: imagine every site as an SL sim Ina

Alpha Auer: Zany: That is what i was thining too

Sean18 McCarey: 3d will no doubt become a part of the internet just like video is now

Razor Briggs: o ok cus im reading in on it im new to all this but im REALLY interested in gettin in on this

Math Gazov: open and different

jesz Murakami: again

Morgaine Dinova: Lem: Gareth holds the bunny, and the bunny does the typing 🙂

Ina Centaur: / reminds me of the Flash revolution, math ;-P

Gareth Ellison: if 3D can be abstracted then everything else can run at lower layers of abstraction

Sean18 McCarey: web pages as they are are still by far more effecient for much of what the web does

Zany Lane: I would imagine that 3D virtual web sites could be as different from netscape as netscape was different from gopher.

Gwyneth Llewelyn: Ina: old article:

http://gwynethllewelyn.net/article103visual1layout1.html

(shameless plug)

Gareth Ellison: if i visit amazon, then i may browse 2D text

Math Gazov: not if you count inmersion Sean

Gareth Ellison: or i may look in 3D at the products

Sean18 McCarey: 3d customer service replacing simple chat

Lem Skall: jesz just made a good point: how much trust do you have in an av that holds a bunny, and I added, what gender is the bunny?

Morgaine Dinova: Gwyn is quite shameless :-))))))

Morgaine Dinova giggles

Gareth Ellison: the bunny is trustworthy

Gareth Ellison: leave him out of this!

Gwyneth Llewelyn: aye, that’s me 😉

Lamont Helvetic: some people have too much interest in a bunny’s genitals

Scope Cleaver: Is that the elusive white rabbit everyone was looking for 🙂

Gareth Ellison: enough about the bunny

Ina Centaur: ew

Lem Skall: oh, it’s a pink bow so must be female

Gwyneth Llewelyn: haha Lem

Morgaine Dinova: Oh, follow the while rabbit!

Zany Lane: Where do you see this going in the next 6 months short term, Gareth?

Zany Lane: Not the bunny, osgrid

Gareth Ellison: Zany – i see the codebase becoming highly stable and feature-complete

Gareth Ellison: i see the bunny eating a few carrots and hopping around

Gwyneth Llewelyn: haha

Math Gazov: 🙂

Alpha Auer: 🙂

Zany Lane: Is it conceivable to consider teleporting from the main grid to osgrid and back?

Scope Cleaver: Did you invite people at LL to work on your code on their free time? 🙂

Math Gazov: depends on LL I think

Gareth Ellison: 2 approaches, 1 legal and 1 illegal

Lem Skall: when will it be, let’s say “commercial” for lack of a better word? so when do you think it will really have an impact as a VW?

Zany Lane: Lets just think legal

Morgaine Dinova: Gareth: how come that your rate of progress is severals orders of magnitude faster than LL’s, it seems? 🙂

Gareth Ellison: legal option is to have LL implement 3rd-party interfaces

Math Gazov: never Zane!

Gwyneth Llewelyn: hehe Morgaine 🙂

Gareth Ellison: illegal option is to use SLProxy and blatantly violate TOS

Zany Lane: I’m not ready to be banned (yet)

Math Gazov: freedom is legal???

Gareth Ellison: i have asked one of the lindens to allow a test on that

Gareth Ellison: thus far no response

Lamont Helvetic: so.. would there be the potential for a sim, with customizable avatars, to run independent of teh grid? for a single company to use for meetings?

Gareth Ellison: yes

Zany Lane: Otherwise, you just give a -loginuri argument to the client??

Math Gazov: I think reverse engineering is more efficient morgaine

Gareth Ellison: right now you can do that

Ina Centaur: lamont, i think running a sim separate from the main grid is already possible

Gwyneth Llewelyn: from LL, you mean?

Morgaine Dinova: Well, items we develop in SL belong to us, or at least don’t belong to LL. But avatars would be a grey area — best to avoid trouble and create new ones.

Ina Centaur: but that’s boring. and you don’t need SL for that.. there are plenty of other 3d worlds to use for the nonconnected version

Afn Bade: is the protocal for the generic metaverse published yet?

Gareth Ellison: afn: an out of date version on my personal webserver

Gareth Ellison:

http://sehll.garethnelson.com/coolstuff/gmmp.txt

Ina Centaur: SL obtained their avatars from Poser 2, iirc

Gareth Ellison: correct typos

Gwyneth Llewelyn: aye Ina 🙂

Sean18 McCarey: items are not transferable from LL unless LL allows them

Gareth Ellison: as for exporting – sadly, TOS disallows it

Gareth Ellison: it’s trivial to do, but against the rules

Morgaine Dinova: If they *belong* to us (as LL’s advertising/FAQs say), then they can hardly withhold them from us.

Zany Lane: Do you see any commercialization of sims on osgrid with sim owners selling land?

Gareth Ellison: Morgaine: it’s tricky, they can forbid use of their servers any way they want

Afn Bade: link for the generic protocol did not work

Lamont Helvetic: thsi seems to be what copybot was made for.. if it is yours, take it with you.

Sean18 McCarey: belong to you for the exclusive use on SL

Gareth Ellison: Zany: yes, it’ll happen

Chase Marellan: The intellectual property belongs to us. That doesn’t mean you don’t have to recreate it somewhere else.

Gareth Ellison: copybot is exactly what i started to ponder using for this purpose

Morgaine Dinova: Gareth: they can’t forbid the use of our clients though.

Sean18 McCarey: how do you get copybot

Gareth Ellison: run a copybot on the osgrid server and do not allow it to leak

Gareth Ellison: for obvious reasons

Zany Lane: Does opensim keep up with the client updates from LL?

Gareth Ellison: ensure it acts only on consent from the user

Gareth Ellison: Zany: yes

Gareth Ellison: my bot was called ExportBot

Gareth Ellison: due to PR concerns and the TOS the project was dropped

Zany Lane: Where are the gotchas in opensim?

Lem Skall: does LL have any plans to implement a way to download our objects off the grid? so we can upload them into something like OSgrid?

Gareth Ellison: Lem: no

Gareth Ellison: at least not public

Gwyneth Llewelyn: No, not public

Gareth Ellison: Zany:

http://bug.openmv.org

Afn Bade: intellectual property is only good if you can enforce it, as anyone can develop content, the value of IP becomes meaningless…. except for a small percentage of IP

Lem Skall: at least they haven’t said they will never do it?

Gwyneth Llewelyn: lol Lem

Gareth Ellison: no, but not said they will either

Gareth Ellison: and i have a feeling they won’t

Gwyneth Llewelyn: in fact, they have sort of planned to allow people to dump all their content in a private island,

Gareth Ellison: don’t want to lose all their customers

Gwyneth Llewelyn: to make backups

Lem Skall: they should, otherwise it is not validating of those objects being our IP

Gareth Ellison: there’s various options for backing up stuff

Gareth Ellison: but TOS forbids many

Lem Skall: but still within the SL grid

Morgaine Dinova: Lem: they have said that they are not willing to enter an arms race with open-source developers. However, they haven’t said that they won’t start doing nasty things with lawyers.

Gwyneth Llewelyn: Hmm Gareth, they won’t lose customers “yet”… I mean… over 90% of the people want to go where the BIG content is.

Gwyneth Llewelyn: hehe yes, Morgaine

Math Gazov: the ONLy great exclusive capital SL has is the stored content

Gwyneth Llewelyn: that’s exactly their attitude!

Gareth Ellison: they will lose customers if the content can be relocated

Math Gazov: content won´t leak out easily from here

Ina Centaur: @math… also 8+ million users

Gwyneth Llewelyn: well, Gareth, yes, if we’re talking about an OSgrid with 10k servers and over 97% uptime

Gareth Ellison: it would be trivial to move lots of it, but sadly cannot be supported due to legal reasons

Lem Skall: then we don’t really, legally, OWN those objects and that stands against all the hype

Afn Bade: The attraction of SL is other people… events were people show up 🙂

Gareth Ellison: any content copying must be done manually

Math Gazov: 8 million unstable and discontent users Ina

Gareth Ellison: that’s legal

Gareth Ellison: but automated copying isn’t

Zany Lane: So, in reality, this is more like opening up the wild west a bit and the rules and roles will probably be changing over the next year or so.

Flyingroc Chung: 97% uptime is over 10 days downtime in a year

Math Gazov: and users aware of other free chioces if they appear

Gareth Ellison: regarding downtime – with OSGrid once stable i have planned to have downtime one day per week or one day per fortnight for updates

Afn Bade: OVer time open source solutions will win over closed source

Gareth Ellison: this is i feel much more fair than SL’s sudden downtime

Morgaine Dinova: Well they’ll be taken to task for saying that community-created content belongs to their creators, if they actually prevent creaters from taking it. “It’s yours but I won’t let you have it” is indefensible.

Gwyneth Llewelyn: ie. a roadmap?

Gwyneth Llewelyn: oh, btw, are there milestones for OSgrid?

Gareth Ellison: none, yet

Gwyneth Llewelyn: ok.

Gareth Ellison: it’s still in the early stages

Zany Lane: Arent there some 0.4, 0.4, 0.5 goals in the wiki, Gareth

Ina Centaur: @math… there really is no other 3d mmorg that has the freedom of SL.. and that’s one difficulty in code of SL, that it basically allows you to do anything.

Gareth Ellison: that’s opensim

Lem Skall: Gwyn, are you kidding, it’s an OS project

Lem Skall: not a business

Gareth Ellison: there’s an opensim roadmap

Gwyneth Llewelyn: hehe Lem

Gwyneth Llewelyn: the best OS projects *do* have milestones 😉

Gareth Ellison: OSGrid is just an installation of opensim and OGS

Gwyneth Llewelyn: even some of the worst… hehe

Afn Bade: the best OS products work lol

Ina Centaur: yes, when would it be possible to own a sim without paying for the setup price equiv to buying a computer or two..

Morgaine Dinova: OSGrid is to OpenSim, as efnet is to ircd.

Lem Skall: in my experience they have milestones only they are pseudo-run by companies like eclipse and IBM

Zany Lane: I think Gareth said today.

Afn Bade: will there be any support for AI for the OS grid?

Gwyneth Llewelyn: Well, yes, Lem …. like Mozilla… or MySQL… you’re right on there 🙂

Gwyneth Llewelyn: Afn, you can develop it if you wish 🙂

Math Gazov: AIs will be born FROM open grids

Ina Centaur: AI support could come from externally

Ina Centaur: i.e. similar http_request

Morgaine Dinova: Well not doubt USAF will be implementing Skynet on it … 😉

Gwyneth Llewelyn: hahah Morgaine!

Gwyneth Llewelyn laughs

Lem Skall: what happened to Gareth?

Scope Cleaver: Mmmm Skynet

Morgaine Dinova: Gwyn: why are you laughing? ;-))))

Scope Cleaver: Tara! 🙂 Hi

Math Gazov: think he crashed

Gwyneth Llewelyn: welcome, tara5 🙂

Gwyneth Llewelyn: I think that our host crashed…

Lem Skall: hi tara and ty

Afn Bade: It would be nice if there was some explict protocol for AI. so you could turn AI off if you want on the grid.

Gwyneth Llewelyn: Well Afn,

Zany Lane: This opensim thing sounds very interesting.

Razor Briggs: can someone help me with the OpenSim program

Ina Centaur: afn, what *sort* of ai? o_O

Math Gazov: maybe LL boicot 🙂

Gwyneth Llewelyn: the cool thing is that you can implement it if you wish

Lem Skall: tara, don’t go!

Lem Skall: lol

Morgaine Dinova: Hey, you’ve got to have AI before you can turn it off

Scope Cleaver: I agree Morg

Zany Lane: I think Gareth said go to irc.efnet.net to the #opensim chat channel for help

Gwyneth Llewelyn: mm hmm

Ina Centaur: … ai=artificial intelligence or another tla?

Scope Cleaver: in fact you could argue it’s no true AI untill you *can’t* turn it off

Gwyneth Llewelyn: 😀

Lem Skall: wb Gareth

Morgaine Dinova: Scope: hahahah — there’s some truth in that 🙂

Tara5 Oh: not going just thought I should have my own seat!

Gareth Ellison: apologies, crappy DSL

Gwyneth Llewelyn: Ah, Gareth is back!

Ina Centaur: … why would you need ai to run a sim? o_O

Gareth Ellison: AI for running a sim?

Zany Lane: Hide the beer, Gareth is back.

Gwyneth Llewelyn: Ina, to allow avatars that are pseudo.intelligent and shoot at you 😉

Scope Cleaver: ONce you get AI up ask it to fix the grid plz

Gareth Ellison: ah, NPCs

Math Gazov: a sim with AI would be alive, no need to “run” it

Morgaine Dinova: The bunny will get tipsy

Gwyneth Llewelyn: mm hmm

Gareth Ellison: just use a bot

Gwyneth Llewelyn: yes, a bot.

Boshemia Vieria: to play God…

Ina Centaur: right, so the bot doesn’t necessarily ahve to be hardcoded into sim software

Gwyneth Llewelyn: Funny how SL, deep down, looks like IRC 🙂

Ina Centaur: you could just port it out via http or xml/rpc

Flyingroc Chung: Does OSGrid require MySQL, or can it use a different DB backend?

Gareth Ellison: actually, that is one feature i pondered

Gareth Ellison: there’s 4 different DB backends

Ina Centaur: yes, s.a. Oracle

Scope Cleaver: Which of them scales best?

Gareth Ellison: MySQL

Gwyneth Llewelyn: (it was once 🙂 )

Lamont Helvetic: hmm… what is teh SL equiv of DCC?

Gwyneth Llewelyn: is Oracle free for Linux? 😉

Gareth Ellison: i doubt it

Gwyneth Llewelyn: hehe it used to be… for a short while

Morgaine Dinova: It used to be free, I very much doubt that’s changed

Gareth Ellison: last i checked it was commercial

Gwyneth Llewelyn: hmm Morgaine, I think people were appalled at how slow Oracle was ;9

Morgaine Dinova: It’s always been commercial, but the free Linux version was a marketting hook.

Gwyneth Llewelyn: and you know how Oracle is so pesky when they get ‘compared’ 😉

Gwyneth Llewelyn: yes.

Flyingroc Chung: Oracle used to have very high latency

Ina Centaur: google uses oracle iirc 😉

Morgaine Dinova: Gwyn: hahahaha

Gwyneth Llewelyn: They do, Ina? Hmm 🙂

Morgaine Dinova: Oh dear

Gareth Ellison: could i ask if anyone has on-topic questions?

Ina Centaur: oracle vs ms sql vs mysql… oracle wins

Zany Lane: Thank you very much, Gareth. This has been very helpful and I will definitely look into osgrid.

Razor Briggs: ya how do you use OpenSim lolol

Tara5 Oh: Oracle not free for linux can download install and run it but need to pay to license it

Gareth Ellison: www.openmv.org

Gareth Ellison: read the wiki for instructions

Morgaine Dinova: Well Orcale ban publication of comparative results for Oracle … so yeah, wouldn’t surprise me if they’re dreadful.

Ina Centaur: yes, oracle is *not* open source

Razor Briggs: i am doesnt make sence to me

Lamont Helvetic: is there an “opensim for dummies” page?

Ina Centaur: … just like how the stable and upbeat stuff tends to not be o_O

Lem Skall: how well is the architecture of OSgrid described on the website?

Gareth Ellison: if anyone wants to discuss technical details come on IRC or email me

Gwyneth Llewelyn: Razor:

http://www.openmetaverse.org/wiki/OpenSim_Build_Mono

Gareth Ellison: the website needs fleshing out a lot

Gareth Ellison: but i’m always willing to answer questions via email or over IRC

Gwyneth Llewelyn: yes, Gareth, it would be nice to give the -loginuri URL…

Gareth Ellison:

http://osgrid.org:8002

Gwyneth Llewelyn: thank you 🙂

Math Gazov: Gareth, why don’t you set a blog?

Ina Centaur: @gareth, re: $10/month dedicated server?

Gareth Ellison: set a home sim

Gareth Ellison: if you have an account, go to the site and edit your OGS profile

Gareth Ellison: then pass that -loginuri and explore

Ina Centaur: home sim would still not be $10/month… unless you live in idaho and have super cheap electricity bills o_O

Lem Skall: we do need to open a new accout, right? can’t use an SL av

Gareth Ellison: yes

Gareth Ellison: register an account on the site

Gareth Ellison: LL haven’t opened their user database to us 😉

Quanta Torok: For those of us without technical skills, does your project accept financial donations?

Gareth Ellison: i will have to look into paypal donate buttons etc

Flyingroc Chung: Are yo hiring?

Gareth Ellison: any funds for donations would have to go to the open metaverse foundation

Morgaine Dinova: Ohio Genealogical Society must be the wrong OGS … 😉

Gareth Ellison: the board for the OMF are still not finalised

Ina Centaur: is the open metaverse foundation a registered nonprofit?

Gareth Ellison: they’re still seting up some of the financial and legal things

Ina Centaur: as would the omaho genealogical society ;-P

Ina Centaur: *omaha >.<

Razor Briggs: ok whats the irc?

Ina Centaur: efnet

Gareth Ellison: efnet #opensim

Gareth Ellison: once the OMF board is finalised i’ll ask them about donations

Gareth Ellison: thanks for the reminder 🙂

Razor Briggs: lmao theres noone talking on the irc

Gwyneth Llewelyn: they’re all on OSgrid, Razor 😉

Gwyneth Llewelyn: hehe

Gareth Ellison: what’s your nick razor?

Razor Briggs: MUAHAHAHAH!!!!

Ina Centaur: lol

Gareth Ellison: at times the channel is idle

Ina Centaur *poofs

Gareth Ellison: yeah

Razor Briggs: johnpablo

Razor Briggs: onirc?

Razor Briggs: you said hi to me

Razor Briggs: lmao

Gareth Ellison: well, i believe this talk has covered all that’s reasonable to cover – any detailed queries please direct to email or IRC

Gwyneth Llewelyn: hehe

Gareth Ellison: [email protected]

Lem Skall: Gareth, is Nymphomaniacs Paradise another opensl related group?

Gareth Ellison: i’ll be disconnecting from LL’s grid before this bunny claws my eyes

Gwyneth Llewelyn: and I hope you’ll do more ‘promotion events’ for OSgrid!

Math Gazov: anyone wants a transcription of the meet? please IM me

Gwyneth Llewelyn: thank you very much for doing this, Gareth

Lem Skall: yes, ty Gareth

Gareth Ellison: lem: that’s a nudge nudge wink wink group

Math Gazov: ty gareth

Gareth Ellison: a friend invited me into it

Lem Skall: 🙂 must be a good friend

Lamont Helvetic: hey, teh nudge nudge, and casinos, have vested interest in opensim

Quanta Torok: Thank you Gareth!

Gwyneth Llewelyn: hhe Lamont, indeed

Math Gazov: we could have regular meets about this

Razor Briggs: [email protected]!T

Razor Briggs: sorry XD

Math Gazov: to know what’s going on

Gareth Ellison: ok, in all seriousness – i will be disconnecting now to do some work

Gareth Ellison: i might organised a meet on OSGrid next time 🙂

Gareth Ellison: oh, search groups for OSGrid and join

Gareth Ellison: if i do organise any meets i’ll announce them there

Razor Briggs: im in lmao

jesz Murakami: bye been intersting!!

Gareth Ellison: thanks for a great discussion and good night

Lem Skall: bte all, I’ll take off too

Flyingroc Chung: um, how to I log on there agin?

Gwyneth Llewelyn: use -loginuri

http://osgrid.org:8002

Gwyneth Llewelyn: mind you, their login server is down 😉

Flyingroc Chung: lol

Flyingroc Chung: ok

Gwyneth Llewelyn: Well FR, you have to understand that OSgrid and Deep Grid,

Gwyneth Llewelyn: are about at the same level as LL was in 2001.

Gwyneth Llewelyn: It works.

Gwyneth Llewelyn: It doesn’t work flawlessly:)

Dnate Mars: yes, because LL has gotten it to be so flawless right now too 😉

Gwyneth Llewelyn: haha

Gwyneth Llewelyn: aye, Dnate 😀

Flyingroc Chung: so, hm

Flyingroc Chung: I thin I better start getting better at C#

Gwyneth Llewelyn: LOL

Flyingroc Chung: Ineed it for my new job anyway

Gwyneth Llewelyn: hehehehe

Gwyneth Llewelyn: also, opensim runs on .NET 😉

Gwyneth Llewelyn: so your hmm new employers will appreciate it 😀

Flyingroc Chung: lol

Flyingroc Chung: last I looked though, libsl was a mess

Gwyneth Llewelyn: hehe

Flyingroc Chung: granted, that was a whle ago

Gwyneth Llewelyn: I can’t comment, I never managed to get it working

Gwyneth Llewelyn: Yes, same here actually

Dnate Mars: .NET, ugh

Gwyneth Llewelyn: hehe Dnate

Scope Cleaver: Take care everyone, nice meeting.

Gwyneth Llewelyn: well, it runs on Mono too,

Flyingroc Chung: .NET isnt too bad

Gwyneth Llewelyn: allegedly 😉

Gwyneth Llewelyn: bye Scope :))

CC BY 4.0 Open Source Second Life – The Geeks Strike Back by Gwyneth Llewelyn is licensed under a Creative Commons Attribution 4.0 International License.

About Gwyneth Llewelyn

I'm just a virtual girl in a virtual world...

2 Pingbacks/Trackbacks

  • Interesting; very interesting indeed. Let’s hope that these people build in proper governance tools right from the outset 🙂

  • Hmm, actually, two questions: (1) how on earth will it be possible to have a Linden-dollar-like currency on an open system like that; and (2) assuming, in five years’ time, that everybody’s using the OpenGrid, how on earth will Linden Lab make any money? I’m sure that it won’t have returned its venture capitalists’ money by then.

  • Good questions! After the recent tumult and chaos on SL’s economy, it’s more than clear that people need to trust whoever runs the currency 🙂

    I’d imagine that the only solution to that is using a third-party micropayment system.

    As for (2), the answer is simpler. When companies release their previously licensed software as fully open source, they make money from services instead. Here is a short list of what LL would very likely be providing:

    1. Hosting services. After all, they will still be the entity on the planet with more experience in running a 12,000-sim grid. Most new customers will definitely be more willing to pay LL to host a server for them than to blindly trust a bunch of kids running servers behind their ADSL modems.

    2. Interconnection fees. Here will be the largest “cash cow” for LL: you’d be able to run your own grid, but if you want to access the uncountless Terabytes of content on LL’s grid, you’ll have to pay for the privilege. And pay quite a lot. In fact, I can seriously expect that things like OS Grid would need to pay up to US$200 per month for each sim they wish to interconnect with LL’s own grid…

    A typical example on how this works are the roaming agreements between mobile phone operators, or the peering agreements for Internet service providers. In those cases, the amount of payment will depend on how huge your own network is. The larger you are, the less you’ll pay — if you’re the same size as LL, and have as many sims and registered users, it’s likely that LL won’t demand any payment — since half of their own user’s content will be stored on your grid, not LL’s. This is how these things work.

    In effect, the huge “Metaverse Service Providers” will very likely agree to interconnect between themselves for free, while reselling interconnection for a huge fee to the smaller grid operators.

    In the same manner, to guarantee that your sim name (or even avatar name) is unique across the grid, you’ll be able to pay a monthly or annual fee for “name registration on the metaverse”, to be provided by LL. LL will be much like Network Solutions — not an “exclusive” operator, but the one that ultimately runs the “root asset servers” for the World-Wide Metaverse.

    3. Technical support. If you are running your own grid, but require support and training, LL will gladly provide it — for a price!

    4. Features. Many will obviously contribute to the open source community and benefit from any new features that happen to be created that way. However, some companies have specific requirements and will require outsourcing that work to LL.

    What all this means is that LL, as a company, will probably be happy to remain the same size and with the current running costs, but will suddenly have access to a large number of pure services at almost 100% profit margins with a way larger grid and user base… and believe me, having much higher margins on some selected services will make their venture capitalists very happy. Specially because some of them are very familiar with open source models 🙂

  • Interesting replies. As to the currency, what kind of third party micropayments model had you in mind? Do you think that the OpenGrid people are going to start an “OpenDollar” service, designed to be able to be used as a currency by any virtual world provider, and, indeed, any website? And wouldn’t such a service also, de facto, be a central repository for avatar name standards: i.e., the client will send the OpenDollar server its username and, encrypted, its OpenDollar password, and will also send the sim server its username (the same as the OpenDollar username), and its local grid password (which would be different); the OpenGrid server would then check with the OpenDollar server whether the person connecting has a right to use that username, and will log the user into the OpenGrid servers only if it does.

    As for all of the other services that Linden Lab could make money from in an open-source world, it strikes me that, once the grid has been open-source for a few years, enough for others to catch up, Linden Lab will not have any particular advantage in providing any of those services over any other company who wishes to do the same. People might not, after all, want to connect to the LL grid if there’s another grid to which they can connect for free (and on which land prices are much lower), especially if, by then, Linen Lab have allowed people to make local backups of all of their objects (and, in any event, just imagine what a cross-grid copybot would do). After all, how much profits do the people who run the DNS servers really make?

    Would the venture capitalists be happy that they have paid Linden Lab to create an opportunity to make profit in a particular way, which opportunity is equally given to any potential competitor from a certain date in the future?

  • I don’t think that there will be an “OpenDollar” system (although I’m pretty sure many will try to do it!) since the issue is really about how to trust them… but there are a lot of micropayment systems available anyway. Amazon, for instance, has just launched one that seems tailor-made for Second Life (well, it shouldn’t surprise us that the founder of Amazon is also a venture capitalist for LL…). But there are a lot of them around – all backed up by RL companies under jurisdictions that regulate their operations.

    Note that it would be quite easy to create an OpenPayPal or something similar — but nobody did anything like that, for the same reason. Would you trust a kid running a payment gateway running from his home, using his own bank account? 🙂 Well, most wouldn’t, but then again, we trusted Ginko… 🙂

    As for LL’s change of business model… the answer is obviously “yes”. First, although registering a top-level domain costs the registrars a few cents (US$0.15 I believe), we’re talking about hundreds of millions of domain names. Even small national registrars for a national top-level domain, with a few hundred thousand domains (which cost way more than that) are able to pay their running costs. They don’t need to sell licenses for BIND (the most popular DNS server) or hosting websites! Remember, ultimately, LL might only be a team of 10 system administrators and a handful of programmers running the “root asset servers” — if everybody else is running their own sims.

    The other thing is medium-term vs. long-term plans. Imagine that tomorrow IBM launches their own grid with 12,000 OpenSim servers. Well, what would the running costs of that operation be? Why would IBM be cheaper than Linden Lab? LL is not really doing “speculative pricing” on their servers just to Make Money Fast™ — an illusion that comes from the fact that a low-end webserver just costs US$100/month or so to run and maintain, and so LL is making a huge profit. This is simply not true! To ensure the same level of commitment and development than LL, you have to add the salary costs on top of the service being provided, and remember, there are just a handful of people world-wide with the necessary know-how to handle a SL grid. Almost all work at Linden Lab 🙂 So salary costs will definitely make an “alternative grid” more costly than people can imagine (in the olden days, people also tried to create an “alternative DNS system” to escape the costs of registering domains, but as the success of that grew, the running costs were too high, and the “established DNS system” was simply run more cost-effectively).

    The image people have is of kids running their own sims at home and interconnecting them, and not requiring labour costs. That’s all fine for “experiments”, but most kids don’t have 5 Mbps upstreaming bandwidth at home, which is what you need to get 50 or so avatars running from your server. Not even small companies have that much upstreaming bandwidth — ADSL/cable connections usually have between 128 Kbps to 1 Mbps, and that barely allows 1-10 avatars to connect to your home sim, on a good day.

    So what will possibly happen is something like the suggested “Open Grid Foundation”, which will get donations to host a large amount of sims on a co-location facility, and, since they’ll be tax-free and use volunteer labour, they’ll be able to run a “cheap” grid for quite a while. This, again, doesn’t scale well — although you can start with a few hundred servers that way, at some point, the running costs will rise above the ability of the Foundation to draw donations to pay for the co-location costs. This will mean that they’ll have to be selective in what projects they accept (like the FSF does), or use another revenue-earning model.

    On the long-term, however, you might be right. Someone with unlimited starting funds might launch their own grid, and create a fantastic and unheard-of business model that allows them to offset the costs of running the grid, without charging the whole price to the end user. This will very likely be a combination of sponsoring and ads. And as we all know, the biggest investors on ads and sponsorship on the Web come from sites with mature content and casinos 🙂 So the first mega-cheap grid will very likely not be used by corporations or universities that will dislike the idea of having cheap sims sponsored by your local sex shop.

    The other possibility would be someone like IBM, Sun, Microsoft, or whoever comes next, to launch a mega-grid, not with 12,000 sims, but a million. Running a grid hundred times larger is not hundred times as expensive, but much less — and this would mean saving costs (quite a lot, in fact), and effectively being able to successfully compete with Linden Lab in the same business area.

    The question is, who would invest now in a million-sim grid? It would require close to a billion accounts (not active ones, just registered ones) to be profitable. Again, I find it quite unlikely that anyone would really invest in such a business model just to be able to place sims in the market at a lower price…

    The last way to effectively compete with Linden Lab would just to use more advanced hardware. Let’s take an example. We have heard from LL that they use quad-CPU servers, and that they run four sims per server. Well, on the transcript above, you might have noticed that Gareth commented that you could run in theory around hundred sims on a single server. The only issue in this case is dealing with the massive networking and CPU load if all sims are full with eager avatars, all drawing an average of 100 Kbps. OpenSimulator does not even have a physics engine, and no assets to keep track of, so naturally it’s much lighter.

    Still, this is just part of the story. One could employ an IBM mainframe running a thousand virtual Linuxes, and a couple of Gigabit Ethernet cards on it. In fact, a single IBM mainframe with 64 CPUs could run the whole of LL’s current grid, with the same performance. And it costs perhaps hundred times less that LL’s whole grid! Sun has also equivalent hardware that could do about the same. So that is a reasonable alternative for someone to start up their own grid. It certainly has its own problems — I’m not underestimating them — and high running costs, but the simple truth is that, in terms of hardware, it’s a far cheaper solution. That’s why IBM and Sun and others are still manufacturing mainframes — for these kinds of environments, they still beat grids of interconnected PCs.

    So, assuming that someone is seriously going to try to replace the multiple-PC-grid by more advanced hardware, there is a way to compete with LL on costs only. However, hardware is just part of the story — bandwidth costs are the same, for a ten-thousand-PC grid or a mainframe running ten thousand virtual servers.

    Still, there are more alternatives. For instance, a single sysadmin is pretty much able to deal with a grid with up to a hundred servers or so — allowing 400 sims. So small “grid providers” would probably manage to survive thanks to very reduced overhead and salary costs. They don’t even need offices. Imagine that Anshe, who runs around 600 sims, would now be able to buy her own hardware, get it on a Chinese data centre with Chinese funding, andhas just an engineer overseeing those servers. She might be able to offer sims at half the cost or less. However, as soon as more and more people buy more and more sims from the AnsheGrid, she’ll run into the same issues as LL.

    In fact, the big question here will very likely be: “do you want to connect to LL’s main grid or not?” If the answer is “yes”, LL will always have a way to charge interconnection fees (or allow only Premium users to “roam” across grids). If the answer is “who cares about LL’s grid?” then there are other solutions: Multiverse, for instance, is quite more cost-effective for someone like Anshe that has her own team of content producers, 3D modellers, programmers, and engineers. She might simply just launch her own virtual world and forget about LL.

    No, for me, the most important thing is to ensure that if LL is removed from the business arena (mostly because of lawsuits and Government pressure), there are ways to go on with a “Second Life” that doesn’t require LL’s presence. Cheaper costs are actually not so important for me; if IBM takes over where LL has left (they have far better lawyers to survive any “attack”), I want to be able to transfer all my assets and L$ to them — and it doesn’t matter much to me how much I need to pay for the privilege 🙂

  • On the OpenDollar

    Ahh, I’m not so sure that it is such a bad idea, you know. There’s a good reason that Linden Lab started using their own currency, rather than using PayPal for transactions: it works better across international boundaries, and works better for very small transactions.

    With PayPal, one has to pay them for each and every transaction, and pay further to convert one currency into another. Transaction values are very difficult to calculate internationally when different currencies are involved: think, for example, how comparatively few people purchase goods from eBay from foreign as opposed to domestic sellers, especially for low value items.

    One also needs a currency that scales down better, since one Linden dollar is worth less than half a US cent, and is therefore far more viable for ultra-low value transactions (and even more so versus the British pound: a single British penny is worth about L$5.)

    People may well not trust an anonymous fellow from Brazil running OpenDollar from his garage (or yacht), but that does not mean that it is impossible to invent a trusted model: after all, people trust Linden Lab.

    Suppose, for example, that OpenDollar were to work this way: a serious group of people, using open-source software, create the necessary architecture, and invest in the necessary hardware, spread across numerous national jurisdictions, with multiple redundancy set up such a system such that it has a permanently fixed number of OpenDollars: say O$100,000,000,000,000. They do not run a currency exchange themselves, but instead act as a sort of OpenDollar central bank: they start with all the OpenDollars belonging to them, and auction them off in chunks of millions to other institutions that would act as the currency exchanges or individual banks, and would sell them to the public. They would then loan OpenDollars to financial institutions trading in OpenDollars at a variable interest rate, determined by a committee of the OpenDollar board, the interest being charged in OpenDollars. The long-term profit would come from re-selling the OpenDollars retrieved by the loan interest (the value of which would be maintained by the necessarily limited number of OpenDollars in circulation; the number retained by the OpenDollar foundation would be carefully monitored to ensure the maintenance of the value of the currency, and they might even buy them back in some circumstnaces).

    On alternative grids

    That is a very interesting response. I shall look forward to the day when there are grids on mainframes, since, given what you have written above, that certainly seems to be the way forward.

  • Pingback: Ecommerce Lounge - Todays top blog posts on Internet Marketing - Powered by SocialRank()

  • A.T.

    oh, such a juicy topics 😉

    BTW, OpenID seems to be forgotten during recovery from blog fault, eh?

  • Damian Poirier

    Porting content is VERY difficult. You can use glIntercept to capture the mesh of your objects, but this is not prim based data anymore so the slviewer won’t show it. Further more it has no skin or script attached to it, nor inventory.

    As these are significant aspects of content you’d pretty much have to recreate all content from scratch.

    I’ve never been impressed with LL’s permissions system, nor the economics of ‘owner’ selling vrs ‘creator’ selling.

    What struck me half way through was, “How can this be OS grid if it’s based on .NET?”

    I’d like to see a sports arena sim where the players are renderd and the ‘field’ but non of the spectators(ghost viewers). How many spectators could such a sim maintain? What would the lag increase per spectator be compared to rendered AVs?

  • Oops A.T…. my apologies. I’m trying to get it back!!

  • Damien: presumably, LSL could be used to port content simply by capturing all the co-ordinates numbers and object properties, etc.? It should at least work for full-permissions objects, even if LL have introduced copybot-defeating algorithms for everything else.

    Why don’t you like the idea of owners of objects being able to sell them? I thought that the trans/no trans permission dealt with that in any case.

    As to .NET, there’s nothing to stop that being opensource – there’s always Mono 🙂

  • it’s Ok by now – seem to appeared again. this is try for it.

  • hmmm, tho not Ok pages – clicking “next page” in bottom returned me into top of blog :-/

  • Math Gazov

    We’re missing here a main point:
    independent grids are a fertile ground for experiments, with low overhead costs and no angry customers they have that luxury.

    The future grids must have a different architecture than LL’s model. Instead of fixing a CPU to serve a number sims, they’ll redistribute processing power in response to avatar density, through paralell processing. In other words, if you visit a sim, you share part of your CPU to process that environment. The SETI project used paralell processing years ago, why couldn’t it be applied in the metaverse? With proper algorithms, that should greatly decrease server overload and bandwith demand, and get us closer to the 1CPU’s avatar dream.

    The metaverse must apply the web 2.0 core principle: more users automaticaly strenghten the grid, the inverse of overloading it.

    The first team that gets such a grid working wins the jackpot, no matter if they’re a bunch of kids.

  • All true, Math 🙂 The importance of the OS Grid should really not be underestimated. Just because it’s nothing more than a toy right now, it allows research, investigation, testing out models, see what works and what doesn’t — while not interfering with the LL-run grid at all. The good ideas will necessarily be “pushed” into SL-at-large sooner or later; as said, LL has people watching closely about what might come out of it.

    Still, there is a fundamental flaw in the overall SL system, and the OS Grid will not address it immediately. SL simulators were designed to work similarly to a web server: they have an amount of finite space and allow a finite amount of users. Massive distribution exists on the Web because people can easily copy a whole site on a second server and balance the load among both. If you need to support more users, you just keep adding servers. The Web, being a stateless protocol, is very good at this sort of copy-content-and-paste-it-on-another-server-to-split-the-load.

    SL, however, is far from stateless! Thus, a different architecture will have to emerge, and we’re far from even imagining how that would work. Dynamic allocation of CPUs on a massively distributed system is way beyond what SL is able to do right now.

    [email protected], and its successor, Boinc, work at parallel computing. Under this scenario, the data to be processed is simply split into “chunks”, and each CPU (on a user’s home computer) gets a chunk and processes a bit of the data, and then all gets assembled back again. A similar model is p2p networking for sharing files: you just split them up in chunks and download each chunk separately from different servers running on people’s homes.

    SL does not work well under this model. Or, if you wish, part of it is actually similar, but not the critical bits. For instance, currently, textures (well, all assets really) are spread across the 12,000 simulators. In theory, your computer could open 12,000 connections to each simulator and get the textures directly from each of them; this way, overall, the simulators would have less work, since the “empty” sims would be helping out spreading the huge load of streaming textures.

    However, this doesn’t scale well. With 50,000 people online simultaneously, the sheer number of open connections required for this model to work would exhaust a server’s ability to handle them (connections are limited to a certain number; this can be tweaked, but it’s not “infinite”; also, you can tweak them on the servers, but what about the clients running on a computer from their homes?). What Linden Lab devised instead is placing a proxy server for assets on each simulator, and making the client’s computer just connect to the proxy server instead, while on the background, all textures required for a particular sim are funneled from all over the grid and stored locally. So while the assets are totally distributed over the grid, when connecting to a sim, a local copy of all the textures you need (around 100,000 for a regular sim; more, of course, when avatars with attachments start logging in) will be ready for download as soon as your teleport finishes.

    If you look at your texture cache in your computer, you’ll see that they are anywhere between 10k and 100k in size (some are way bigger). I didn’t compute an average, but this basically means that a whole sim could very well take about 1-10 GBytes to download. You don’t get all the textures in a sim, however — things like occlusion will prevent some faces to be displayed, and it’s worthless to send textures that will never be seen. All this is rather cleverly done. But as you know, your local cache can be set only to 1 GByte, meaning that a single sim can simply exhaust the size of your own computer’s cache. Walk across a dozen of sims, and it’s the end of your computer’s disk space 🙂

    Now, how can this be solved? Right now, as you said, figuring an alternative solution will most definitely “win the jackpot”. All we know is that things are way complex — more than the average programmer imagines — and that LL’s solution, while certainly not perfect, was not so silly at all. I just think that they grossly underestimated user mobility (ie. people are hopping from sim to sim at all the times) and, of course, the amount of people that will gather on events.

    I imagine that a first way to deal with this problem of massively distributing the grid would be to separate three aspects of a simulator’s job: tracking users, streaming textures, handling physics. The reason why most MMORPGS can get far more users per CPU is because they don’t need to stream textures, and to a degree, the physics engine is loaded on the client. So the servers only handle tracking data, which is the least demanding issue on the sim. The problem is that the “tracking engine” in SL is starved for CPU resources in a busy sim, full with textures of all avatars that are dropping by, and with the physics engine dealing with all the movements and bumping into walls and other avatars. If the solution is to separate the three things between different hardware — I don’t know. I can imagine that “dynamically” tracking users — not unlike what OpenCroquet does — might help to make it more distributed. Assets can be offloaded into separate systems — in fact, weren’t it for the costs, LL could simply put all textures on Amazon’s S3 distributed infrastructure, and just handle tracking & physics. In fact, even with the current costs, it might pay out (I haven’t made the calculations). The physics engine, of course, is a major culprit. Havok 1.0 simply cannot handle this. Havok 4.5 might help, as well as splitting the task of handling physics between the client and the server. However, this would mean that the client could not remain open source any more. Plan B, of course, is to use an open source physics engine running on the client — many exist — but that is definitely a major development for many years…

    So, in the short term, I fail to see how things could be made “better”, except for delegating the responsibility of the asset server and the other centralised servers. This will naturally help a lot — even allow parts of the grid to keep running when a “local” asset server dies — and the nice part of those changes is that they can be performed on a “live” grid, over time. And offloading the stress of streaming textures to a separate infrastructure would also help a lot. The rest, well, requires some genial breakthrough never attempted before — coming from the “bunch of kids” or LL 🙂

  • Here’s my blog on why we need Micropayments — and in Lindens.

    http://secondthoughts.typepad.com/second_thoughts/2007/11/ten-reasons-we-.html

    I see more clearly now what a World Transhumanist Society *conspiracy* all this hacking and open-sim stuff is : )

    Morgaine is in full regalia here ROFL.

    Well, transhumanism is pretty much on hold, waiting for nanotech, which got delayed by Bush and vested interests. Annoying

    Actually…it’s not that this sectarian and esoteric ideology is “on hold” — it’s just that it can’t find adherents because it is too rigid and too wacky. Bush is hardly responsible for “delaying nanotech” (?!) and the concept that “vested interests” are to blame sounds like some kind of nutty conspiracy! Maybe nobody will pay for Morgaine’s nanotech sandbox *shrugs*.
    http://slrecord.typepad.com/the_second_life_record/2007/11/a-jointly-arriv.html#more

  • And they continue to strike, again and again http://paymentguy.blogspot.com/2008_02_18_archive.html

  • Pingback: Blog Xemelê » Metaversos: o ‘hype’, a bolha, e o futuro()