More information and some thoughts on LL’s new, work-in-progress, virtual world

My take on all this

Like many people in SL, I once was under the terribly mistaken impression that whatever we asked for in feature requests could be done with relative ease. The implications and complications of working on an old codebase with a lot of old content whose functionality must not be touched and could very easily be broken dawned on me when two things happened in my life: (i) I ended up working in RL for a small company that depends on such software, (ii) I started frequenting LL’s official user groups.

Really, these two things were eye-openers; they woke me up to the difficulties of professional software engineering like a bucket of ice-cold water on the face. I found out through personal experience that all this “LL are evil, Oz is an *EXPLETIVE DELETED* that goes out of his way to piss users off, etc” kind of talk is a supertanker of horseshit. I’m terribly sorry, but there’s no polite way to say it. If you’re still yapping on about how “LL steals your content” or about how “LL won’t listen to its users” or how “LL is maliciously trying to annoy its users”, you need a reality check. And if, even when presented with facts, you still choose to not let them spoil your story… Well, I guess there’s nothing I can do for you and, if you really think LL’s people are going out of their way to piss you off, I can only wonder what you’re still doing in SL.

Once again, maintaining Second Life is not easy. It’s old. It’s monolithic. There are many things in there that are so inextricably tied to the very way it works (such as the 256x256m regions) that changing them could break everything. Many of the conventions that were made back then were valid in 1999 and 2000, but no longer relevant or desirable now (such as the default avatar mesh or the default camera offsets).

Linden Lab, in my opinion, chose the correct course of action. They’re not killing Second Life (after all, it’s their most lucrative product so far), as Peter Gray wrote to several bloggers (including Inara Pey) on the subject, and they’re not stopping its maintenance and development. What they are doing is leverage their experience from Second Life to make a better platform. Hopefully with a better scripting language, better in-world building tools and (please!) support for lag-busting procedural textures.

Personally, I’m not fearful for Second Life. I’m certainly going to give the new virtual world a go when it’s ready, especially as my interest in virtual worlds has now ceased being that of an amateur. Whether I’ll like it more than SL or less will remain to be seen. As for our legacy content… Well, if and when the time comes for LL to pull the plug on SL, there will always be the option to migrate content that can’t be transferred to the new platform to OpenSim, with the proper arrangements regarding permissions. After all, there are people out there who have a presence in other virtual worlds besides SL.

.

See also:

.

Shortlink: http://wp.me/p2pUmX-A2

11 thoughts on “More information and some thoughts on LL’s new, work-in-progress, virtual world

  1. At a guess regarding backwards compatibility…
    – Would you give a wild guess at what proportion of SL’s present content could be transferred into a new world using macro-style automation?
    – Basically the question is, how much of that work might be mechanical ‘translations’, and how much of it do you think would be too abstract and requiring higher level functions?
    – In absence of knowledge of the new worlds, would you rather guess that from-scratch build in them would probably be easier and better?
    Bruce Thomson in New Zealand.

    Like

    1. Even if you addressed these questions to Ebbe himself, you wouldn’t get an answer – even he himself wouldn’t know, as things are certainly too fluid at the moment. I’m semi-flattered that you think I might be able to answer any of them, but I’m not privy to any of LL’s technical workings and I must say I don’t feel comfortable being seen as an LL insider.

      Like

  2. It will be interesting to know what graphic engine the Lab chooses for the new world. Will it be voxel-based like HiFi? Or, will it be compatible with Allegorithmic substances, as is the case with the Unity and Unreal engines? It would be awesome if LL adopted the Substance Engine. Or, C# for scripting…

    Like

    1. Re: voxel technology, I’ve no idea. Allegorithmic? No idea. What I do know is that Allegorithmic’s Substance Designer is proprietary, it’s Windows-only, and so I’m not sure how attractive that would be; you can find tools for making bitmap textures on all platforms (Windows, Mac OS X, Linux). I do think, of course, that support for procedural textures would be hugely beneficial. As for C#, I’ve no opinion on that… And no idea, either.

      Like

      1. Actually, Allegorithmic products are not Windows only. The tool to make procedural textures, Designer, is available for both Windows and Mac. Painter is still in beta. The engine, which is required to stream their compressed format (one of best benefits in adopting Allegorithmic’s technology), is available for Windows, Mac OS, Linux, Android, iOS, PS4 and Xbox1, and that is required only server side. I don’t see the fact that the technology is proprietary to be a limiting factor server-side. Even the graphic engine LL will adopt will be proprietary, just as Havok engine is for SL.

        I am not sure if substances require a client-side texture renderer, as that may be a problem for 3rd party viewers. In such a case an arrangement could be made with Allegorithmic to provide a free or low-cost library.

        Like

        1. Still, tools are needed for all operating systems (including Linux and Android). As for the proprietary nature of the engine, there’s one disadvantage: vendor lock-in. If the engine’s vendor goes under, the customers are in a rather uncomfortable situation.

          LL also uses Kakadu’s proprietary technology for the textures. And the insecure piece of crap called Vivox for voice chat. I know that LL has secured licences for its TPV devs.

          Like

          1. Sure, vendor lock-in is a disadvantage but unavoidable in some cases. Unless LL decides to develop all the components of the VW on its own, including the graphic engine… 🙂 As for Allegorithmic’s technology, I see that in addition to regular texturing, not as a replacement.

            Like

          2. The way things are, I don’t think developing everything from scratch is an option. And yes, providing Allegorithmic as an additional texturing option is what I had in mind too.

            Like

  3. One thing I would like to see in any future LL virtual world: protection of our identities. Leaving behind our content will be painful enough; I’d like to see a promise from them that our names will be available to us if we choose to participate in the new world, and will not be available to anybody else.

    Like

  4. That was always one of the things Rod Humble emphasized, the right to Anonymity was and should always be a fundamental part of Linden Lab, regarding Sl.
    I do wish the same will apply to the new product even if I’ve my doubts about the goals and audience it will aim towards.
    Still, let’s be fair, all is to soon to guess and even if i’m prone to the most terrible scenarios for now all i care is for Sl and the joy it gives me.

    Like

Comments are closed.