Taking Second Life to the next level

by Pixeleen Mistral on 03/11/06 at 8:57 am

What does it take to be the Web-3D market leader?

by Stef Wade

Stef Wade ponders how to level up

It is no secret that the Second Life client puts quite a strain on many computers. The german blogging scene is discussing some technical insufficiencies which can be found in Second Life. Some go so far as to say that as soon as big companies with experience in MMOs begin to realize how much money there is to be made in virtual reality, Linden Lab will soon lose its market leadership. Sony and EA are mentioned as potential rivals to Linden Lab.

But I believe that it will not be the superior 3D-engine or better client performance that will decide who will be the market leader in virtual reality applications. Instead, this will be determined by the other factors:

* External Access: Can I manipulate my avatar and inventory somewhere outside of SL? Can I do so using standard tools – like an ASCII-editor? Can my self-made (or bought) software access my avatar? Can I backup my inventory to a DVD and put it into the company’s safe?

* Open Servers: Do I need to have my Sim hosted at Linden Lab or can other servers be integrated into the metaverse? Are there interfaces to other VR-providers? Can I decide myself, how much computing power I want to dedicate to my sim? Servers should be able to implement different features. Today there are regions with “Restricted llPush” or those rated “M” or “PG”. Server admins should be able to decide which features they want to support. I could imagine servers which support more avatars in one Sim, but restrict textures to three basic types: Building-texture, clothes-texture and plant-texture.

* Integration of the Old-style Web 1.x/2.x: I need to read my e-mail, websites, RSS-feeds and other things in the virtual world. Can I edit them there? In order to spend my working day productively in SL, I need the same possibilities I already have on my RL PC.

* Integration into the Old-style Web 1.x/2.x: It should be possible to fetch any kind of data from “regular URLs” and store it there. I want to be able to host inventory lists, textures, clothing, whatever on my servers on the “old web”. HTTP and WebDAV are already available to do that. This would also ease the performance problems, if SL-sim-servers would only store references to the real location of the data. That data could be hosted on web servers, which use optimized and tested techniques for high traffic sites (load balancing, high availability, …) Also, the “external access” would be much easier.
SL-shop, web shop and real-life bookkeeping could work on the same database. SL-transactions would be recorded straight into the regular bookkeeping. And if I host my inventory, my buildings and my avatar on my company’s server, then daily backups should be no problem, right?

* Integration into “real life”: SL needs to blend into RL like the “old internet” has done. The border to virtuality must not be recognizable. At least not more than entering a car or staring a web browser. SL, web and RL will blend into each other, and only if the difference is not noticed anymore, virtual reality will be a success.

It is not easy for Linden Lab to give up the monopoly they have in SL. But as far as I can see, it is the only possibility to be successful in the market. An open system will invite competition to the LLs own court, but it will be the only way to prevent the marketplace from splitting into many small pieces.

Linden Lab will make money as specialists with a lot of know-how in SL. As a VR-hosting-company, so to say. If they want to be the sole supplier of virtual reality, they will suffer the same fate as CompuServe, AOL and MSN: As closed platform none of them was able to gain leverage over the others. When the open competition Internet came along, none of them was able to sustain their old business model. Today, all three are regular content providers which have integrated more or less into the open system, the internet.

Also, manufacturers of closed computer-systems (Commodore, Atari, Apple, Sinclair,…) had to recognize that their decline began with the dawn of an open system, which integrated competitors. Except Apple, none of them is one of the big players anymore.

Resistance is futile…

13 Responses to “Taking Second Life to the next level”

  1. Urizenus

    Nov 3rd, 2006

    Good one. But we should post little bios of some of these new writers, Pix. Who is this guy?

  2. Prokofy Neva

    Nov 3rd, 2006

    I’m one of those non-tekkies who consistently argues against this standard tekkie open-source vision because I think there’s absolutely nothing in heaven and earth that says you HAVE to repeat evolution/design/historical experience every time. Any argument that tells you condescendingly, with know-it-all assurance, that “it has to be this way just because that’s how we did it last time or those were the mistakes last time” is simply not persuasive to me.

    Just because AOL is viewed as tekkies as a dinosaur that failed because it wasn’t open, doesn’t mean that today’s AOL-type thing will fail. It may not. You have to keep your mind open about the development of Web 2.0 or you yourself will be overtaken by events.

    AOL qua AOL isn’t such a failure, either, as hordes of people still use it for email and swear by it, and it still has significant market share. Ditto Apple. Just because these so-called struggling dinosaurs of Web 1.0 had a tough time doesn’t mean they didn’t survive, and who knows, maybe they are the very ones positioned to make it big with the metaverse even while everybody else whines about proprietary rights and refusal to go open-source.

    My problem with the open-source concept, which has prompted me to say constantly that “open-source=closed society” is the following:

    1. Only privileged tekkies who understand all the features and issues outlined here get to reap the rewards of the soi-disant “open source” mash-up style software — so it’s not open to start simply due to the comlexity of its use. It’s open *to that class of people only*. What *is* open for democratization and participation by the mass public and even intelligentia (who just doesn’t happen to be tekkies) are models where a company like an AOL or a tripod makes a basic template for dummies, and the dummies then use it to focus on CONTENT rather then FORM (technology).

    So I draw a firm distinction between open SOURCE and open USE and I favour open USE if it is created by having proprietary sources. I don’t fetishize the openness of a source, if all it does is open it up for USE only by a handful of lucky tekkies. I don’t whine about proprietary rights when then create a vast vista of open USE.

    That is what has made the web, the blogosphere, and podcasting what they are today — not smarties using technology and only talking endlessly about their technology, but dummies using the dumbed-down technology to make more interesting CONTENT.

    Open-USE versus open-SOURCE. Start this discussion. Have it today. And stop reciting the mantras. And open USE for me means easy hurdling of technological battles, simple user panels, and even if it means payment, low cost.

    2. The proliferation of thousands of DIY dungeons made by open-sourcerers will lead to an archipelago of egos. This Balkanization of the 3-D web is merely a closed society hiding under the trappings of an open society. It’s merely thousands of tekkies whining about the few bigger companies being too authoritarian simply wishing to become *as authoritarian* as they are — that’s all lol.

    They are only interested to grab open source for themselves, and then make things that are…closed. Run by them under authoritarian regimes that brook no dissent and have no common universal values. At least with the Lindens, as bad as they are, we have more or less a TOS and a kind of benign if weak federal government. We won’t have that when elitist Snowcrash tekkies make evil utopians and make people dependent on their whims.

    3. I constant see in this article the words MY MY MY MY. But this is a shared, contiguous world. Um, just what is it you want to be saving on to your hard drive? My land? My products? Or just yours? See, that’s the issue. If you make an open-sourced, extractable system, then all my labour is rendered null and void. If I make content, you can extract, copy and sell it elsewhere without my permission. Even my terraforming is something you can copy and endlessly render. So no thanks! It’s not protectionist to point out that a system that sucks the value out of people’s labour and creativity by endlessly replicating it like that creates a real problem.. which is…

    4. Who pays? People aren’t willing to go pay homebrewers all the time; people need a place to start and get oriented; people need other people to meet. So companies with skills and capital will provide this, and they need to get paid to do this. These Medici-like game companies have too much power, but their highly-paid and skilled inhouse game devs cost money, and those people have to figure out a way to make the game pay.

    The new media and the new Web 2.0 will be unlike their past variants. They might contain features of the old-media public movie, with its $7 ticket. They might contain elements of the web that are free like Yahoo mail — with its ads. Most media traditionally needs to live off advertising, as subscriptions alone can’t pay for it.

    One thing is certain about all this: we need many types of disciplines and types of minds on this problem. Just as Web 1.0 could not stay hostage to all those hard-left utopianists and Reason readers and Randians of various stripes of the Well and sequelae, so the new Web 2.0 will have to get beyond the same bunch trying to clutch at it.

    In fact, it’s already bigger, and the fact that there are so many people in WoW and even SL despite the profund distrust and even scorn that many large enterprises feel toward video games and social worlds, it is accelerating beyond them. It’s because so many ordinary people can access them and walk around the tekkies who’d like to serve as gate-keepers to keep themselves employed.

  3. Stef Wade

    Nov 3rd, 2006

    @ Prokofy Neva:

    You made me reread my own writing. Not once did I mention open source. Not once did I say “give SL away”.

    Yes, I am enthusiastic about open source, and I usually log into SL using the linux-client. But I did not request SL to be open-sourced.

    I said that it would be wise to open the technologie (can be on a paid basis!). If the technology will not be open, competitors will need to their own thing. And they might do it better than the Lindens.

    “I constant see in this article the words MY MY MY MY. But this is a shared, contiguous world. Um, just what is it you want to be saving on to your hard drive? My land? My products? Or just yours?”

    Both: I want to save mine, and I want you to save yours. Because if LindenLab will lose to some competitor, “then all YOUR labour is rendered null and void” – including your terraforming! You can’t even make a screenshot, if you are not quick enough!

    Do you actually expect medium-sized companies to invest into their presence in SL, and if that fails in “Life reloaded” and then maybe in “EA-Life” or in “Vaio-Ego” all over again? You are right: Someone has to pay. But right now, most companies that invest in SL are so big that what they spend there is just peanuts for them. Once these investments are secured by an open infrastructure, more (and smaller) companies can afford to follow.

    I did not say that any of the companies I mentioned are failures. I just said that they had to change and adopt to an open system.

    BTW, you only pounded on claims I did not make. So, I guess that you either recognized my love for open-source “between the lines”, or you projected your fears into my writing…

  4. Nobody Fugazi

    Nov 3rd, 2006

    Stef – pay no nevermind, your article was good. Your thoughts on extending vr< ->web2.0< ->web1.0 is something I have been considering as well.

    Prok brings some good questions to the table, but the truth is that the answers are pretty much apparent. You just have to pay attention. :-)

  5. Prokofy Neva

    Nov 3rd, 2006

    Stef, your article is a good summary, that’s not at issue. But it’s fine to take issue with your ideas, which are merely the same things that every tekkie who walks into SL says, sooner of later.

    Yes, I realize the whole Internet is based on open source, and Ll uses open source too, but my questions remain legitimate.

    When you say, “Open Servers: Do I need to have my Sim hosted at Linden Lab or can other servers be integrated into the metaverse? — that IS open-source. How did you think I would host my own server, yet connect that server up to LL’s server farm? With…what?

    I would hope that whatever happens, SL will be sold with the stuff on it.

    Yes, licensing is something they might do, too, and make people pay a bundle to be able to do that host-your-own thing. Essentially the emulation of that on private islands now is being priced out of the market for individual proprietors with no resources.

    No, I didn’t “project,” I just read the same set of theses I’ve read 100 times before from tekkies about SL, over and over again. Google “open source” or “host server” with Second Life on Google.

    Nobody, yeah, the answers are there, and they involve “having tekkie utopian socialists devalue my land investment and creations,” I realize that. And, no, that’s not progress. Merely in the name of training a legion of script -kiddies to work for the next big thing created by bigger corporations.

    Yes, your Lenin said he would sell them the rope…

  6. Stef Wade

    Nov 4th, 2006

    One last time, and then I will be taking Nobidy Fugazi’s advice:

    * “Integration” and “connecting” and “working together” do NOT mean open source. The more you write, the more I get the impression you don’t know what open source is and what it is not. MS-IIS can server pages into the internet just as the Netscape server did, just like apache does. No problem for the internet.

    * The point which lead me to stop this discussion is: Open-Source does NOT mean communism. I run a business myself. But this is probably not the forum to discuss that.

  7. Tomas Hausdorff

    Nov 4th, 2006

    Yowsa: sorry for the extra posts. I received a strange “connection refused” error when I tried to post, and I guess this is the result- please delete the spares copies, thanks ;)

  8. Tomas Hausdorff

    Nov 4th, 2006

    It is one thing to suggest that SL should have an open / published API for the client, and another to say that it should be possible for folks to just install SL Server on their box and launch their own “mini-Grid”.

    Firstly, how does a user who pays for content (E.G.: an avatar, a suit of clothes, a nifty gadget) carry that item reliably from independent server to independent server? What if the servers have different “rules” that disallow features and functions the person running the avatar paid for? Would you trust “Joe Anonymous” to host your inventory and transact your Lindens?

    Maybe we solve these problems by making each server isolated. But if each server is an isolated “world” with its own set of rules and very limited interconnect between them, how is this a “metaverse”? It isn’t: its a bunch of isolated pockets.

    What people calling for private/independent ownership of Second Life sims often do is call out how successful the Internet has been with privately owned HTTP servers. True, but browsers on the web carry no “presence” with them. Other people browsing a web page that I’m on can’t “see” me. They don’t have to know what I’m transacting, and do not view any form of stateful representation of what I’m doing. The content I view and access on one page does not need to be carried over to the next page and faithfully reproduced, stored, and executed there. These are non-trivial differences.

    Personally, I like the idea of opening up the interface to permit different “views” into Second Life- perhaps customized clients, externalized mechanisms for building and managing content and the like. I am much less fond of the idea of splintering the world itself into thousands of privately owned and isolated “shards”.

  9. csven

    Nov 4th, 2006

    SL’s tools are effectively open-USE, yet we’ve been hearing complaints about content creators since I can recall. If someone points out that it’s more than SL’s tools that are involved, namely an external tool like Photoshop, then there’s the obvious open-USE GIMP available to once again level the barrier to entry. But the anti-content creator spiel continues unabated, doesn’t it?

    I sense some projected fears myself.

  10. Piett Hesse

    Nov 6th, 2006

    LL will eventually learn the web 2.0 way: start with the API’s. If they give public API’s access, people will praise them, just like did to Flickr. It provides tekkies means to make people use those services in a easier manner. The world is complex. Knowledge is necessary to take care of complex things and translate them to simpler language. See how Flickr photos can transparently be integrated into blogs. This is done because tekkies used the API’s to build code to allow people to see the photos in their blog platform using a simple mouse-click.

    Stef, Prokov is no tekkie nor he knows about what OS (open-source) and FS (free (as freedom) software) are. Maintain a discussion may be tricky because of miss-concepts.
    Proprietary software usually match with closed source software.
    Prokov, these terms are a bit complex to be debated without some tekkie knowledge:


    The SL grid lives over a free software structure. Well, i am writing this post on a linux machine. I run SL on the Linux client. When Fugazi says:

    “Prok brings some good questions to the table, but the truth is that the answers are pretty much apparent. You just have to pay attention”

    …i do feel like posting a smile too :-)
    The inventory (etc) backup can be done without OS. A simple public API call can manage it.

    Take this open source:
    # Piett Code followed by a GPL license…
    class Mixin
    attr_accessor :a, :b
    def initialize(a, b)
    @a, @b = a, b
    def >
    puts “Yep!” if @a > @b
    test = Mixin.new(2,1)

    Now build or run that open-source code.
    Or take a closed binary file:
    ./Mixin (Mixin.exe for Windows)

    If i register that code with a, say, GPL license, you can use, change, copy, give it. In this case, someone could fork LL’s SL and create a 3L. But it would have to be GPL’ed too. And this makes the world evolve in a community way, and may the best win. Take profit out of it ? I can collect the same fees like LL’s SL. But if someone creates a 4L, better then my 3L, i may start to loose clients. Or not.

    I’m too tired now, and maybe you are not getting to my point. The thing is, all these are *points*. You can use the strategy you want (LL in this case). This discussion exists for more than 30 years, and still no “consensus” was found.
    Take Microsoft Windows and Torvalds Linux Kernel. Windows is closed source proprietary software, but they have API’s (Win32 API). Linux kernel is open source free software. I can build up a Piett Kernel based on Linux in 10 secs. But i have to say that it is based on Linux Kernel and print the license. I can sell it. I can give it away. I am free. Here, if quality is not assured, the product dies before it was born. Plus, it is given a chance to never die. Hard to read and understand others code ? You bet. Terrible. Highly complex, sometimes. Just for geeks. But as i said, the world is complex. You get features by adding complexity to objects.

    Passing all this to SL. Have you heard about SETI@Home project ?
    This is a distributed computing project. Berkeley Univ. had powerful servers choking radio signal data to find, simplifying, patterns. They were investing a lot to maintain those servers. Suddenly someone created a form of distributing the heavy task of iterating through the signal among the millions of home/enterprise computers, so that Berkeley servers only get the results to verify their authenticity and post/store. This made the processing power raise exponentially. This made the community gather around this project and embrace it, so that from single home users with no astronomy knowledge, to IBM/HP/…/Google like companies, there’s a huge will to make the project go forward, and even see who chokes more data. Who wins ? It’s a win/win situation.
    LL can allow the creation of small, controled, personall grids, without loosing a bit of their business. They for sure will increase it, as i can serve a small local community on my small grid. All this controlled by a central LL grid. And finally all this with proprietary *or* non_proprietary software for the:
    - Main Grid
    - Client or Satellite Grids
    - Clients

    This, i bet, is the LL SL future. Then they can mashup services in and outside. I bet LL is brainstorming a lot to define *the path*.

    Finally, this is a very nice discussion we could have inside.
    Stef, Prokov, Csven, Fugazi, Tomas, shall we meet inside to have a chat on this ? We could even build a small conference named “SL::Next_Step.InformalTalks”, invite some Electric Sheep, MillionsOfUs,…, LibsecondLife, Linden guys/gals. Who’s on ?


  11. Nobody Fugazi

    Nov 16th, 2006

    Ahh, Prok. It must be easy for you to label people you don’t like in the hope that people will follow that label.

    Perhaps you’re the one who needs the label. ‘The Disingenuous Labeler’. Yes. I think that fits. :-D

  12. alien_genius

    Apr 14th, 2007

    My response may be a little late, but I still am going to comment on the original post and the discussion that followed.

    I am a programmer, webdeveloper and 2D and 3D graphics designer, with more than 25 years of experience. My focus when creating products is always on usability for non-technical end users, including extremely clear documentation.

    I have been wandering around in SL for a while and have paid attention to the technical aspects, but automatically I also am forced to think about the (non-technical) user experience.

    As a consequence I have to agree with the general points made by Steff Wade. Eventhough he is not speaking abut “open source” (which means source code included, free to modify and redistribute) he speaks of open technologies. The best term would be “open standards”, like all the internet protocols are open standards, like HTML and XML are opne standards, and so on.

    The open standards are what makes the net tick. Nobody has a monopoly, there are no hidden features, and probably most importantly, it makes interoperability between vastly different architectures possible. I.e. The same webpage will display on a PC, a Mac or a Linux box. The same emails can be sent between these systems, etc.

    And these open standards do not stand in the way of proprietary code, commercial applications, or any other business operations. The actually help facilitate them.

    Integrating the web, email and ftp into Second Life is almost inevitable, because it brings the biggest portion of internet activity into the realm of SL. Fortunately Linden Labs is also aware of this detail. In 2005 they mentioned integration of the (open source *wink*) webbrowser FireFox into SL : http://secondlife.com/vote/get_feature.php?get_id=5

    SL also has created a connection between in-world Instant Messages and email, so you can receive your Instant Messages in your email in-box and so your responses via email are received as Instant Messages in-world. All this will probably be expanded into a full fledged email system.

    Since there are quite a lot of open source FTP programs for all the major platforms, it should not be too difficult to integrate FTP as well.

    Next to that, Linden Labs has now open sourced its viewer application : http://www.internetnews.com/dev-news/article.php/3652796

    The other important issue that Steff mentions, is “external access”. In slightly more concrete language, people need to be able to make their textures, sounds, prims, gestures, and all other assets with some program of their choosing, even when they are OFFLINE.

    The reason is that this will make things more accessible to non-technical people, eventhough Urizenus seems to think otherwise. There already are a lot of people who know how to work with a great variety of content creation programs, although they can not really be called technical people. They are creative end users.

    There already are some plugins popping up for use with Blender and 3D Studio Max, so prim building can be done offline, and then imported into SL. Personally I hope that this development continues.

    One more note about user created content in SL. It is nice and dandy that the end user license agreement states that content creators retain copyrights of their creations. However, witout the option of saving your own creations offline on your own computers, that particular clause is almost meaningless.

    By allowing people to export their own objects, it will further reduce the distinction between the virtual and the real world.

    Personally, I hope that the use of open standards and integration of open source products will continue, and that Linden Labs will have the business sense to give end users (technical, non-technical and commercial) more leverage. Because THAT is what can make or break SL.

    I hope that instead of petty namecalling, future posts will be about the actual content of the posts.

  13. stefan.waidele.info

    Jun 23rd, 2007

    Mein Posting im SecondLife Herald

    Die englische Übersetzung meines Artikels Erfolgsfaktoren für SecondLife wurde im SecondLife Herald veröffentlicht: Taking Second Life to the next level
    Das erste Flame ist auch schon eingetrudelt. Aber die Finger habe ich mir noch nicht verbrannt …

Leave a Reply