Slashdot Mirror
Second Life Open Sources Client
An anonymous reader writes "Just noticed that Second Life released their client under the GPL today, and that they're up to 2.4 million users. Article says that 15% of users contribute scripted objects."
← Back to Stories (view on slashdot.org)
There are various open source objects in the game. The "gnubie"-Store is actually not "GNU", but it's often full mod/copy.
It's usually set to no transfer to prevent abuse (but there are plenty of full mod/copy/transfer freebies around, probably houses, too!)
Unfortunately, you can always take away permissions, meaning that you could take that FLOSS house, mod it, remove the next-owner-can-transfer/mod/copy permissions, and basically make it a closed source thing.
Dunno if it hit it yet but yesterday it had almost 25.000 citizens 'in-world' it was floating around 24.500 ..
And Teleporting and or moving around was almost impossible
It has had a lot of media attention lately, all over the world.
Wish they would upgrade their servers, to handle the load of users. Or at least get a EU Server farm.
The official announcement from Linden is on the Second Life Blog.
- You don't know how to maintain a station wagon either!
Current numbers:
Total Residents: 2,434,170
Logged In Last 60 Days: 883,536
Online Now: 13,150
That is right now, right this second as I post this. The highest I've seen the online now number is about 23-24k, and once it gets over 20k shit really does hit the fan.
As far as 15% contribute scripted objects. Perhaps that's 15% of the real active user count, but it sure as hell isn't 15% of the 2.4M. Scripting in SL has a steep learning curve and many people who do building in SL avoid scripting because it is such a pain.
If you think education is expensive, you should try ignorance -- Derek Bok, president of Harvard
While it's not a 'combat game', there are areas that are combat zones. Your avatar can 'die', but I believe it just spawns your avatar somewhere else. (I don't do the combat stuff, I've just heard about it.)
"If you make people think they're thinking, they'll love you; But if you really make them think, they'll hate you." - DM
It's 11.7MB zipped for Windows, plus 45.7MB in additional libraries nessecary to compile (such as Boost)
#!/bin/csh cat $0
Well, wouldn't it be profitable to make a client that just keeps notifying the server of new objects that sell well in the world? I'm not too clear on the crafting process in Second Life but I imagine it takes resources.
The way I understand it, is that there is no "crafting" system per se, but users create things outside of the client, and then upload them to the SL system. Users can then set flags in their creation that makes it non-copyable, non-transferable and/or non-sellable. Therefore a client that creates items perpetually would not give a user any advantage, as items can already be copied at will. There is no rules in SL (except in player created environments), as it is more of a virtual space than a game. This is what I've understood of it, I haven't played it myself so this is only second-hand information.
The reasons that MMORPGs need to "protect" the client don't apply to SL, or are easy to avoid. There are basically three reasons that MMORPG systems have problems with "unauthorized" clients:
Assuming bots aren't an issue, there should be no security-related reason for any MMORPG not to open-source the client.
Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
3D Objects are created entirely inworld. Animations and textures are created outside of the client. Object flags (Mod, Copy, Transfer) can be toggled according to the existing flags and your creator and owner status on the item.
I think we concur...nobody has seen SL and been "like, wow"
You're defending an interactive world that already exists, which is the www. everything you said could have been pasted from 1995 about the web in general.
The ability to "create" things is interesting, but i'm not really sure how viable it is as an ongoing model. scripts layer and then...bog. Many 3D engines and game platforms already allow for custom content - and have nice followings..with much nicer effects.
Realtime nicely rendered avatars with voice chat is closer to the masses via 360 than SL. Plus, in those games, you can do much more interesting things, imho. Streaming music to avatars instead of people is no big win. With Shoutcast (and similar) someone could pump out a singing/dancing avatar vid with real music 5 years ago. Post a VR dancing baby and some music...oh wait...been there.
SL is a diversion, at best.
Applications that use OpenGL or Direct3D are subjected fairly heavily to Amdahl's Law. The problem is that there is a (quite severe) penalty to submitting data to the APIs from different threads, which generally means that all submission to the API is done from a single thread.
If the limiting framerate issue for your title is submission to the API (and for a lot of 3-D graphics applications, it is [warning: PDF]), then you're not going to get any speedup on multi-core systems, and there won't even by a way to improve the latency when typing messages because displaying those messages is what is taking so long.
I currently have no clever signature witicism to add here.
5GB is a number Joe came up with after checking out all sources (servers, clients, libraries, scripts) plus unit tests, packaging tools, and test plans. None of the 5GB includes user generated content of any kind (i.e. textures, lsl scripts, objects, etc.). One can get into all sorts of arguments about methodology (as seems to happen whenever we publish a stat), but the point is, we've got a lot of code.
Today was a small step in the grand scheme of things. As one of the other posters pointed out, the amount of code we posted was a number much smaller than 5GB. But, this will hopefully be an important step in giving people control of their own computers. We certainly don't want you to have to install proprietary software on your computer to enjoy Second Life, and now, you won't have to. Admittedly, there are still some rough edges in a purely open source compile, but that's a bug, not a feature.
Here's my shot at it. 3d in a webpage probably isn't neccessary and to me, it represents tech for tech's sake. But if you insist (CAD, very specialized use, experiemental interfaces) then maybe flash and some XML model (X3D) VRML replacement would suit you. Of course, replacement standards don't really address your question.
... it's just something I've seen attempted in a 3d environment.
What I think the GP is saying is exactly what I think of VRML as bad design. For example, let's say I was going to design Online Phonebook Town. Sigh with me, this should all look familiar to a lot of you, it's been tried so many times. Ok so I might have these key design elements:
- You can run through rivers and streams to interact with a huge phonebook in the middle of a town square.
- When you are looking for a plumber, you go into the plumber building in the town square.
- The plumber building has a huge wrench sign hanging out front (of course).
- You can play MP3s from your computer in your "home" with a huge 3d control panel.
Sounds shiny and exciting, but the actual result is a 3d version of what is already usable in 2d. When people want to call a plumber they aren't really looking for a 3d experience, they want information transparently. When they need a phonebook, they don't want to run over a grassy hill with WASD and see a sky texture. When people want to play music, they probably already have a music player. Who in the world has a collection of MP3s and are just waiting for a VRML virtual town client to come along and play it for them? This feature bloat is an indication of poor direction.
Instead of navigating a horrible 3d metaphor, I'd rather just search in 2d with yahoo. Instead of having a 3d cd player in my "town", I'd rather have Windows integrate with Winamp/iTunes/Whatever. Instead of giving me a 3d desk with WASD controls, models of my "recycle bin", some rehashing of folders as drawers or files as pieces of paper in an 3d application; how about putting more work into the lower level components like the OS or display interfaces? The background music player might work better with a side display on the keyboard or a second monitor. Anyway, nevermind the audio player example
I hesitantly like mature and modern attempts at this 3d metaphor like Project Looking Glass and Compiz/Beryl but so far the best use I see is a flashy demo movie. I'd rather watch the demo movie rather than depend on it. The usability isn't there, VRML was just the start of this continuing effort to get something usable. Beryl is great for making a 2d desktop slightly 3d. And even then, sometimes goes overboard with window wobbling (wtf). The effect is neat but it's kind of like an ice sculpture (pretty and useless).
To sum up my opinions:
1. VRML has a history of creating bad interface metaphors.
2. The 3d metaphor for navigating large amounts of data has mostly been replaced by 2d visualization.
3. Online data interactivity has largely been replaced by AJAX and Flash.
4. Any attempt to organize interfaces together seamlessly should be done with a smart OS not with a bad 3d metaphor for how I work.
5. A full 3d metaphor is useless. 3d icing on a 2d cake is ok, just don't add too much.
A senior developer from Linden spoke at my class a few months back, and he told us about their plans to open source the client, and eventually the server system - i.e., you will eventually be able to use your own hardware to run a SL land server over the internet. The question isn't whether it will happen, it's when. Will it affect their business model? Of course. But according to the developer, the goal is to open-source everything they can, and if the business model has to change, then it'll change.
LL make their money by selling server space. You can't just connect your own server to SL - it has to be one of theirs. The network is closed. All of the PR and astroturfing that's been coming out of LL recently is aimed at getting more people to invest in SL space: the more investors there are, the more the space will be worth. They're trying to drive a homesteading boom like the one that happened in the early days of the Web, when companies started to go online.
On the other hand, this will be a very concrete experiment with micropayments on an effectively wide open information network. If a fully open client with just a trusted third party to handle financial transactions (or maybe ecash) can support a viable information economy through either donations or some form of copyright respect, then it bodes well for a similar "real life" system of micropayments for information and services. Realistically, since most things digital already happen on the Internet, SL will be as real as it gets in terms of the future information economy.
It's likely that Linden Labs is betting on being the manager of the Linden Dollars in the new economy and making their money that way. Hosting server space is a relatively mundane activity compared to the management of the actual money and objects used in SL. Perhaps they will now act as an object ownership repository, basically just keep a hash of every object along with the name of its original creator for the purpose of micropayments to the real creator. Make all objects fully copyable (to respect the reality of information sharing), but let everyone know who the original creator was. Obviously the problem is formally intractable because anyone can modify an object and claim that they're the inventor, but generally market forces will prevent that from happening. Once enough people have seen the original object, they will be able to spot fakes, and since anyone can copy any object, it will be easy to demonstrate the imitations as cheap knockoffs. Payments would just be donations from people who would like to reward the original artists for their creations, and of course for customizations and other services.
I'd also note that Linden Labs has always claimed that SL would be open sourced at some point, so it's likely that this has been planned out quite a bit in advance.
It's CCP, and the concurrent user record is a shade over 33,000. And there are technically two, since there a separate server in China. Now, 33,000 concurrent users converts to some ungodly number of transactions per second. For the load they're handling the game runs phenomenally well under load. From a player's perspective lag can be an issue with some regularity, but we seem to be past the worst of it. EVE is growing and so the server infrastructure is lagging behind capacity. They're buying servers, upgrading the network and tweaking things to reduce the load, but they're still behind. The recent network upgrade seems to have helped a lot.
CCP has done two interesting things to avoid lag. A couple dozen solar systems run on a single blade. High traffic systems have been moved to their own hardware, but the problem comes when a fleet action breaks out on the same node you're on. Their plan is to try and dynamically predict load and move systems off of nodes that are about to suffer some severe lag. In the latest expansion they overhauled the gang system to make it a hierarchical fleet structure instead of just a flat group. I'm predicting that in the next round fleet commanders will be able to designate systems as objectives and rally points. That will let the load balancer start moving other systems off of that node.
Fleet actions really lag the system right now. The biggest I've been in was a 40 v 50, but I've heard of 200+ ships per side. I'm a Battlecruiser pilot. I launch 4 drones and put out 3 missiles every 12 seconds. That's a lot of objects in space when you figure that battleships launch more stuff and there can be hundreds in a fleet action.
The other interesting thing is they recently removed the need for bookmarks to move quickly and safely. Many pilots in alliances would have had thousands of bookmarks. CCP deleted just the so-called instas, and left your other bookmarks intact. That took a relatively small but still significant number of objects out of the database. At the point they did it, even a percentage point or two would have helped.
Let's hope Linden Labs is paying as much attention to their performance issues.
Veteran, Bermuda Triangle Expeditionary Force, 1992-1951