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)
Really, this is a great step towards "Cyberspace" á la Snow Crash. Open Source and, eventually, Open Standars will vastly spur development of this technology.
Linden does not have 2.4 million users, and it does not regularly report how many users it does have. It reports "Residents", a figure that includes people who have signed up for Second Life but never logged in. It also double-counts people who have more than one avatar.
n s_second_life_numbers_and_the_presss_desire_to_bel ieve.php#comments
s econd_life_numbers_thanks_to_david_kirkpatrick.php
More about the uselessness of the Residents figure here: http://many.corante.com/archives/2006/12/26/linde
The only person to whom Linden has reported a count of active users is David Kirkpatrick of Fortune, and as of last week, only 252K people had logged into Second Life twice or more in two month -- the rest were bailouts. This 252K figure, which is a much more accurate reflection of Second Life's popularity, is an order of magnitude lower than most of the press is reporting.
More on Kirkpatrick's numbers here: http://many.corante.com/archives/2007/01/04/real_
Second Life is utterly dated graphically and has a primitive client.
This open source effort is a bid to get the community to do what Linden Labs
has failed to do thus far -- bring their offering into the 21st century.
The clock is ticking for Linden. If anyone thinks that there won't be a better,
more sophisticated and vastly more profitable virtual community within the next
five years, they're either dreaming or they're one of the suckers who has invested
in virtual real estate believing that Second Life has some unique grip on the
concept of virtual communities.
Open Sourcing the client is an effort to cinch public acceptance of Second Life
as the defacto standard in virtual communities. My bet is that Second Life is
dethroned faster than anyone expects. The experience just isn't remotely
sophisticated, graphically rich or slick enough to have staying power.
I downloaded the client a few months back, created an avatar and wandered around:
I felt the experience was primitive, with sub-par graphics, a horrible UI and poor performance
(I'm on a PC graphics workstation with a very fast connection -- that should easily have been able
to handle it). The music was some sort of cheesy new-age MIDI composition, and the character
models seemed like 1990's low-poly attempts at something stylisticly mid 1980's. The character
interaction was poor, there were clipping issues and there was a poor response time with
the environment.
I uninstalled the software within 1 hour.
I'll never log in to Second Life again, and I remain convinced that the contest to be the first
to develop a compelling virtual community is still a wide-open race.
But in terms of statistics, I can assure you that Linden Labs still counts me as a "Resident".
Which begs the question: How representitive am I of Second Life residents in general?
------ The best brain training is now totally free : )
SL has a number of problems. One of them is that the client is well, slow. Framerates of 5 FPS aren't entirely uncommon in some areas. Now instead of blindly speculating, we can look at it and actually tell whether it's just badly coded, or the nature of SL makes it work slowly. This will probably also spur some effort in trying to make it take advantage of multicore CPUs.
Another thing to try would be rewriting the UI. It would be a lot less painful to use if the UI and display weren't in sync, so that when things were slow you could still type at a normal speed.
My personal area of interest would be attempting to provide some sort of way to let SL objects provide a better interface. The sort of interface that can be scripted in SL is very primitive as of now. Being able to make an object with a full dialog with buttons, dropdown lists, a list view, etc would really improve the usability of complex objects.
This should also give a big push to the libsecondlife project, which is also a great thing. SL can be used as a platform for interesting things, such as A-Life experiments. That's another thing I plan to try eventually.
On the Linux side, I'd like to see the integration of something like DCOP, or at least a named pipe to communicate with the SL client. For coding it'd be wonderful to run 'make' and have all the modified scripts automatically sent to SL. Currently this requires an edit, copy, paste into SL cycle.
SL client doesn't really need to be protected.
This isn't WoW, in SL the server takes care of pretty much everything, and the client is practically a 3D web browser. The client is already very unresticted as far as MMORPGs go, you can teleport anywhere you want for instance. Of course you can be banned or not allowed to some destination, but changing the client won't change that.
Even without it being open, the libsecondlife people had figured out enough to duplicate in-game objects. This means that very possibly creators of things that aren't scripted are going to get screwed. But this was always a possibility. It was completely obvious somebody would do it within a few days of trying SL, closed or not.
L$ handling is of course server-side, you can't create them out of nowhere. L$ are only created by LL and then exchanged between residents and bought and sold for USD.
I find it interesting that DESPITE the fact that the majority of posts from "People who know" that Second life is a steaming pile of crap that it continues to grow.
5 8975402950&q=second+life that was done for google.
Go watch the video at http://video.google.com/videoplay?docid=-51827597
The fact of the matter is SL is VERY interesting due to the way in which it's built. It's flexible and the people who run it are BIG proponents of open sourcing everything they can. When you ask them about the number of users they tend to be honest about what they think is real and what are just scripts running. The BS is usually from Trolls.
As for the quality of the graphics.
1. All the content is USER CREATED. Go someplace in SL where people know how to use Blender or Maya and it looks great. Go someplace made by somebody who just learned how to sculpt prims yesterday and it sucks.
2. There is a GREAT live music community growing in SL. The quality is pretty good since you can get up to 768Kb/s of bandwidth to stream your live event.
3. Guess what? The graphics are as good as the clients can handle considering that their primary objective at this point is a flexible world that allows users to create what they want and be scalable.
The majority of people who "crap" on SL (that I've talked to) expect something like WoW. WoW is a TOTALLY different monster. Scripted world, Blizzard created objects...and a much lower age group demographic.
If you want WoW...go play WoW. But don't expect SL to be LIKE WoW.
Of course you could run your own private server, like the Construct in the Matrix. You could do things like the "jump" program and "learn karate". But unlike the movie, you can't carry your guns from the fake fake world to the real fake world.
John
The real problem with SL is one of scalability. In the real world, we work on a combination of peer to peer and server based models; server-based because you have water, power, and communications services delivered to you; peer to peer because your house does not depend on your neighbor's house for anything, and they are effectively equal (even if their sizes are wildly disparate, for example, they both perform the same function.)
In Snow Crash, Stephenson's "Metaverse" was also a peer-to-peer network. It would seem to be highly similar to the web in some ways; links between servers, the capacity for hosting, et cetera. Of course, in Stephenson's world, cheap and plentiful bandwidth connects subscribers (in the form of L. Bob Rife's cable network.)
To make this long story short, we need a distributed architecture that allows you to host your own part of the game world. Monetary transactions between servers would occur in legal tender, and you could have any kind of currency you liked in your game world (if any.) Money transfers could be carried out through any number of services (paypal, egold, whatever.)
This permits as much scalability as you can afford. If you have the money, then you can have your "land" hosted elsewhere; otherwise you put it on your dinky little home connection and only a handful of people can connect at once. Still, this is pretty much the only way to accomplish this goal, and it keeps freedom in the hands of the people.
For a light technology demo version of this, one could add inter-server portals to Sauerbraten. In itself it wouldn't give you the full experience however, as there would be no scripting. Still, Sauerbraten is a collaborative building environment, so it would be interesting in itself.
For something a little more likely to be the future than Second Life, check out Alan Kay (and others)'s Croquet. Croquet is based on Squeak which in turn is a graphically rich Smalltalk environment. Thus Croquet is (or will be - it's in beta now) portable, consistent (Squeak has its own VM which is very consistent across all platforms) and fully Open. Not to mention, it works as I described :)
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
Think of SL as an IRC server that is being run by The Sims. Most people I know in SL go there for the community. The 3D representation gives them a better sense of the person they're talking to, and a different level of interaction.
You CAN do alot for free. There's alot of good free stuff out there. You can also make stuff if you want to. Nobody says you have to buy it. Natural resources? LIMITLESS! Want another box? Just drop one. With SL, the ability to sell goods isn't limited by any sort of ability to produce quantities. It's about design.
If you think education is expensive, you should try ignorance -- Derek Bok, president of Harvard
I think, you're right, but nobody is forcing you to buy anything in SL. You can still participate in the "social culture" without ever spending any money. I haven't. And if you need something, you can simply create it yourself. If I need money to upload a file, I simply sit in a camping chair.
I never bought anything, but created my own skins, textures, objects. I never bought land, because I'm not interested in making money (renting land for a shop) or "settle down" (build a house). I never understood the concept to own a "physical location", a house, in a world which you only experience when you're logged on and roam that world. Why would anyone log on and stay in a virtual home and not visit the many places?
There have to be some people owning land to show their ideas, so other can look at them and experience them. It's like the Web. Most part of the "surfers" just roam the many pages, but a few create their own to share their ideas. No one creates a webpage and locks everyone else out. Well, they surely exist, but they don't matter.
Well, for starters, you've missed that major companies are buying "land" in Second Life and conducting meetings and interviews there. Politicians have conducted town hall meetings there. And Reuters has opened their first cyberspace news bureau in SL. As easy as it may be to mock these bits of information, I think they represent something very important: many companies and services have attempted to be "the metaverse" a la Snow Crash in the past, but what's going on in Second Life is, as klunky and hesitant as it may be, the metaverse actually happening.
For a year I worked for SL's erstwhile competitor, There, which was the one most people were betting on to "win" when both got going. (There made the cover of Business 2.0 and got out of the starting gate with companies like Nike already selling virtual products in-world.) And in a lot of ways There's client technology still kicks SL's ass; the experience is much smoother, even on less high-powered hardware. There's in-world "look" was designed by actual artists, including a former Disney imagineer or two, so when you wander around your eyes don't bleed. There has a sophisticated VR auction system designed by one of eBay's original employees. There accepts models created with GMax rather than a klunky proprietary design system, and ThereScript is based on Lua and is considerably better than Linden's scripting language.
But what Linden figured out that There didn't is that user-created content is king. SL really didn't give a damn if your eyes bled -- they opened the floodgates. Old "Therians" may boggle at my mention of ThereScript, because AFAIK There still hasn't opened it up to users even though they were talking about it when I was there in 2003. (There also had outstanding bugs in the "consumer service" that were going unfixed for months, if not years, IIRC, which were less a matter of technology than politics.)
Personally, I think SL's "under the hood" design is its Achilles' heel, and open-sourcing the client isn't going to help it -- they have a stream-everything model (possibly because their original team apparently came from Real Networks?) and the object system really isn't as sophisticated as what you'd find on an average MUCK server. Someone out there is almost certainly working on what amounts to "Third Life": a design and engineering sensibility as good as There's was (or at least aspired to be), with the understanding of the marketplace and user desires that Linden has. When this happens, that service will be the metaverse equivalent of World of Warcraft to SL's Everquest.
But between all the jokes about flying penises and the ritual mocking of the furries, I think SL is going to prove to be historically important in shaping an "avatar space." Yeah, the idea that a decade from now, it'll be common for businesses to have a virtual storefront in avatar space sounds pretty crazy, and I certainly wouldn't bet on it happening. But you know, in 1994, I'm not sure many of us would have predicted that by 2004, businesses that didn't have a URL would seem to be behind the times.
If the client end of your client server app needs to be "protected", your security model is already terribly flawed. The first rule of client/server app development is simple: Never trust the client.
If you never take input from the client at face value, then you don't need to "protect" it (a war you'll never win, by the way).
Raph Koster knows it. Why other MMO developers have historically ignored this rule over and over again, I'm not sure.
Blizzard has had a terrible track record of violations, however: numerous Diablo 1 hacks, map hacks in their RTSes, up to and including WC3, etc. Frankly, I'm stunned WoW hasn't had any major hacks to date -- maybe they finally learned