Rogue Second Life Server Goes Open Source
While Prokofy Neva, Tony Walsh and others wonder what will happen to Second Life real estate investments once it’s possible to run your own open-source server (i.e., create your own virtual land), someone has gone ahead and done just that. This thread over at the libsecondlife forums describes a Second Life server built by a user, a single sim that can be accessed using the standard client software. As near as I can figure, the server was built by examining the client code (and/or reverse-engineering it) along with the information that is passed between client and server in order to get an idea of what the server code would need to look like. What’s more, an early version of the server code has already been made available as an open-source project. While this doesn’t mean that people can immediately set about creating the distributed metaverse that 3pointD often likes to contemplate (questions such as where everyone’s assets would reside in such a place remain to be answered), it is a step toward a virtual world, solar system, galaxy or universe free from centralized corporate control, one that looks more like the World Wide Web of individually controlled sites than it does a contiguous grid such as exists in Second Life.
The server sounds fairly crude at the moment, with a number of functions yet to be implemented. But it can already handle much of what the official server does, and the number of hackers working on it is growing (though slowly, it seems).
Who knows if it’s possible to grow a robust metaverse out of open-source servers, even if they’re up to speed. But having a parallel version of Second Life out there should at least help speed the development of solutions to many of the problems that plague the application, from lag to UI issues, search and so on, including what seems to be a borked, broken or simply bad asset server problem. 3pointD looks forward to further developments, and would love to hear from anyone working on this or similar projects.
Feeds
[…] 3pointd.com reports that a Second Life server emu is out there, raising the question of what the world looks like when there’s multiple servers to log into. There’s already been some discussion of what happens to the current Linden business model on the day that there’s lots of hosting choices — sort of like a much more sober version of the issues that arose after the CopyBot scandal. […]
It’s a bit less than a personal sandbox server at the moment, no scripting, physics, assets, or grid interconnectivity.
[…] Looks like a parallel world is being created in Second Life’s image. This ‘metaverse’ will be distributed instead of centralized. Now anyone can create land and value will be based on location not scarcity. Anyone with a server will be able to ‘expand’ the universe. In a distributed world Phillip Rosedale wouldn’t be King, instead he would be a simple citizen. I wonder if Second Life will connect their world to the new open world? […]
Prokofy Neva doesn’t raise these issues due to being some evil landlord or gullible swampland investor, but because people want worlds; they want value; they want protected spaces.
Did you really want to live on a web page with 30 strangers, 3 of whom are crashing the page, or recording your personal information to steal from you or stalk you? Seriously. Walker, why are you always so ecstatic about making everything like an open, porous sieve, as if a sieve preserves freedom and creativity and openness? An open sieve merely lets in every flotsam and jetsom.
I suppose somebody like Anshe doesn’t have to worry about Jesse Malthus’ BotWorld which is likely to look like a used car lot in Detroit lol.
Perhaps if enough little scrappy rogue worlds get started, that might flush off some of the hackers and griefers so that civilization might take root better. But by then, they will have figured out how to steal all the content. It will be one long Fifth Avenue with fake Gucci bags for sale on blankets on the sidewalk…
Prok, my feeling is that people who want worlds will tie their bits of the sieve together into little mainlands. We’ll still have worlds, they’ll just be smaller, and more numerous.
The fact is, very few people fly across more than a few sims at a time. The vast majority of travel in SL is done via point-to-point teleporting, usually via search. Functionally, that’s little or no different than navigating a lot of little disparate islands via a Web-based interface or a search function that lists all the little islands. Functionally, I think the change will be less noticeable than it might seem.
Others? It is certainly good to be noted as the majority. Perhaps others can be applied to others.
In as far as it being a ‘rogue server’, well - its a neat trick, but until the actual server code is opened source it will remain shots in the dark. That said, it is a respectably large undertaking of questionable worth… but it may pressure LL to open the source to the server more quickly - but probably not, since it is a standalone. A singular sandbox with no way of intersim communication sounds pretty useless - maybe it is or will be more, but beowulf clusters could make that interesting to say the least. But thats just another opinion.
Oh - and thanks, Prok, for not being too verbose. :-D
oops, sorry, Nobody. Didn’t mean anything by it.
Walker, it stands to reason, however, given the expense of worlds, that big corporations, the dinosaur media that is desperate to revive its dying carcass, will be in a position to spend a lot, house the server farms, and make these worlds BIG and as vapid as a lot of SL is.
And you’re deeply, deeply wrong in your elitist/tekkie/superficial take on the idea that the “vast majority of travel is by p2p2″. Hell, no. It is definitely not. And some day, the Lindens will confess this. Just do the minimum journalistic research on this, PLEASE.
The vast majority of movement comes from the following:
o set home to here
Then, next in line:
o Offer TP to friend
Then, after that (yes, yes, yes, sorry, but this is the reality)
o Use of SEARCH to find a keyword, a place, a store, and TP from the ad
Or even as common, or more common, can’t be sure:
o Use of Map List to pull down sim name and travel to 128/128 mark.
While I realize this is unfashionable and politically incorrect, it’s the case. People do not pull up the map and look around for a place to drop in on . That is, a small, hardy core of those who can really read maps and like them, who like to explore rather than socialize, sure, they use that. Most people use profiles, offer TP, bookmarks, anything but pulling up that laggy map that blocks their whole screen.
People actually like to fly and walk around, they probably do less of it since p2p. Solve the sim-seam, the’ll do more.
I don’t see why we should celebrate the idea of the Lindens’ hand being forced by a bunch of hackers and thugs with bots. Huh? I hope they don’t cave to this.
Sorry Prok, I wasn’t clear. What you describe is exactly what I mean by point-to-point teleportation: teleporting from one point to another after finding the second point by various means, whatever they may be.
Agree with you on the sim-seam problem.
I don’t think the Lindens are in any danger at the moment. As I understand it, the rogue server is pretty crude, as noted in the OP. Just wanted to note a possible future trend.
Nobody says “A singular sandbox with no way of intersim communication sounds pretty useless”… but that is *exactly* what many businesses and educational institutions want - a safe, self-contained world that isn’t in danger of being invaded by furries, Gors and others deemed unpalatable and a distraction from the task at hand.
[…] Es que han sido lanzadas a la Red por uno de los representantes mas importantes de los propietarios en Mainland,y, en lo que a los hispanos nos concierne, los inversores de nuestros paÃses, que no han sido especialmente pro-activos en comprar tierra y abrir proyectos en SL, si que son muy sensibles a estas declaracione, por “el efecto dominó”, me explico. No he encontrado hasta la fecha, ningún analista económico (Expansión, Mercado, etc.) que por su propia cuenta se haya preocupado de hacer un análisis serio del potencial de SL, pero sà que se limitan a copiar y pegar las transcripciones de los medios extranjeros, especialmente los USA. Asà que lo que se consigue con estas declaraciones es ahuyentar posibles inversores. Otra cosa serÃa si se hiciera un contra-análisis. Si bien es cierto que el hecho de que con el futuro (y ni se sabe cuando será ese futuro, pero mañana mismo no, desdeluego)sistema de software abierto, cada usuario que monte un servidor propio tendrá su tierra y sus edificios, no es menos cierto que por un lado están los promotores de “tierra y edificios para todos“, también por otro están están trabajando para desarrollar un sistema de inmersión total en 3D para SL, lo cual fomentará lo que mas les gusta a los avatares: Nosotros mismos. Reconozcamos el chauvinismo que tenemos los usuarios de SL. Nos gusta mejorar nuestro aspecto, poder tener una mejor ineracción con amigos y parejas, y, por lo tanto es ahà donde preveo el mayor futuro de negocio en SL. Skins, shapes, moda, animaciones serán las que mas se soliciten. A estas opciones el tema de la tierra y la construcción les serán indiferentes. Si no tiempo al tiempo, […]
This server (since dubbed “OpenSim”) is rather rough around the edges. I imagine it being some time before this particular project becomes a significant worry of Linden Lab. That being said, it will certainly help some forms of testing and might see some interesting applications for off-grid building areas.
Hah, well, exactly as I predicted; the more information that one allows people to have regarding the client and what it expects to receive, the easier that they will find it to construct something sending appropriate information to said client - in other words, a server.
I know that LL was aware of this at the time of client open-sourcing, and this is precisely the reason why they are making public noises and doubtless private plans regarding, at some point, licencing or releasing server code. With a basic server now, in a year’s time who knows what someone will have produced? Best to ensure that one’s protocol is the dominant one for constructing virtual worlds (who would bother building their own when there is already a client out there and open server code to build on?) and then set oneself up as the best company, with the most experience with the protocol, to go to for a world. Rather that than fight on a playing field that one did not create.
[…] 2 - Rogue Second Life Server Goes Open Source “…it is a step toward a virtual world, solar system, galaxy or universe free from centralized corporate control, one that looks more like the World Wide Web of individually controlled sites…” (tags: cyberspace futurism development platform server source open distributed worlds synthetic metaverse Life Second) […]
Now if that “server” would run under linux! :) Ric
[…] […]
“Rouge” is an inappropriate name for what should become fairly commonplace in the not-too-distant future. Just as one can access any web-server, owned and operated by anyone, through the same browser client ( IE, Firefox, Opera etc) so should it be possible for users to connect to any SIM from the same client ( the 3DBrowser ?)
Shall assets be transferable across servers ? Depends on how we configure matters. When we browse a webserver, not all components reside on the same server … the images could be somewhere else, as could be the CGI programs or the java applets .. so it should ( or could ) be with multi-server virtual worlds.
I think in the future of the metaverse, there should not be an asset server. Each sim should be aware of what is rezzed in that sim, it’s attributes and so forth. But all assets, should be stored on the computer of the person, in some properly documented format for an object (probably some type of xml).
Having a person hold their own assets, themselves, is the most natural way for the think to work. As a comp. sci, I have to drop everything we know already about a thing, and imagine it fresh. And when I imagine various regions, and people who have a lot of stuff, it seems obvious the stuff is with the person, and the region knows only about what is happening within that region.
Whatever we do with federalized simulations, which is what this will end up doing since I imagine folks will want some sort of inter-simulation services - I hope efforts are built towards establishing 3rd party guarantors for various aspects of the simulation. One would involve guaranteeing authenticity of objects, another for simulation behaviors, and yet another for avatar identity.
While I’m not completely cool with copy protection and artificial value (I think it’s incredibly awkward and anti-society) I don’t think we can expect this model to degrade, and even if we had infinite copies of objects without regard to object value, we’ll want to have guarantees that something isn’t malicious if it’s being executed in a “safe” execution space - much the way software is being digitally signed these days.
Like it or not, we’re facing an uber-complex software environment where sandboxing code execution isn’t a guarantee of trouble-free online experiences, and builting a trust model akin to digital signatures would work well if we want to maintain a vibrant online society of this type.
After all, if you do have a pseudo-unique identity in the form of an avatar, and we lose the “protected” environment of a commercially operated simulation such as Linden’s, what sort of free-for-all will we result with? Will desirable people emigrate from that world to free-for-all sims? Things have changed a bit since the old MUD days.
Just my opinion. YMMV.
@ Rickey Moore:
I managed to run OpenSim on Linux (Ubuntu Edgy) yesterday - it runs on Mono. I don’t think I’m a computer genius, I just followed the instructions (start from openmetaverse.com) and I met no difficulties in making it run, then logging in it with the standard client. A funny though mystical experience, one that I would recommend.
…wrong URL: actually, it’s http://openmetaverse.org
also useful are http://www.osgrid.org and http://openmetaverse.org/wiki/OpenSim
[…] Mientras, desde dentro del metaverso, algunos adelantados como Alberto Navarro, organizaban ya grupos de trabajo para aprovechar la apertura del código cliente de SL con vistas a crear un metaverso distribuido: Situación actual. La liberación por parte del Linden del código del cliente ha sido el escopetazo de salida. La comunidad en favor de la liberación del código de Second Life no se caracteriza por su paciencia, y con algunos hackers en sus filas, se han integrado dentro del proyecto de 3pointD en sus objetivos de conseguir un Metaverso Distribuido, y para ello han iniciado su proyecto de código abierto, ó secondlife liberado: libsecondlife. […]