There’s been something of an ongoing discussion over the ever-excellent Metareality concerning the viability / attractiveness of a “new” Second Life – that is, a platform wherein Linden Lab starts over to present something new and overcomes the shortcomings of the SL grid as we know it today.
It’s an interesting – and entirely hypothetical – discussion point. Just how viable would a new Second Life be (if we assume the money was there to develop such a beast), both in terms of Linden Lab’s development of the platform and in people’s acceptance and use of it?
Well, some of the benefits that might come from such a product would be technical; doubtless things like the creaking mass of the asset server infrastructure could be addressed and made a lot more robust / scalable. Potentially the region / sim code could be completely overhauled to both improve stability and remove much of the “band aid” code that has, due to the nature of the platform, had to be applied to deal with various issues and bugs over the years rather than LL being able to dig deep and resolve them at source.
A new Second Life grid could also, I assume, be better geared towards handling the likes of mesh and other capabilities. Similarly, the Viewer could be revamped – and while this may draw boos and hisses – be kept closed, or perhaps licensed, to better control the growth of features and to ensure Viewer and server code remain better integrated.
There might also be the opportunity to directly address issues of accessibility through other means – tablets, web pages and mobile devices.
Social aspects might also be better integrated into the platform as well, for those who wish to use them. These are no to everyone’s cup-of tea, but that’s no reason to exclude such extensions / capabilities.
All of this could be massively to the good; but what about those of us already engaged in Second Life? Are we likely to leap onto the bandwagon of a “new” Second Life? Some undoubtedly would; but many of us probably wouldn’t for much the same reason as we don’t take a deep plunge into existing SL alternatives: we have an awful lot of what amounts to personal investment in our inventories, and if we can’t take it with us, the likelihood is, we aren’t going to go – not unless forced out of SL itself (which might easily see us giving LL the one-fingered salute and disappearing somewhere else entirely).
Of course, losing the current user base (or a good proportion thereof) might be seen as part and parcel of the risks involved in developing an updated platform – after all, with 16K-a-day sign-ups for the current platform, there is opportunity for LL to address initial retention head-on and harness a good percentage of that 16K and so not actually miss those of us who stay behind.
On the other hand, offering a migratory path from “SL 1.0” to “SL 2.0” would obviously be one way of alleviating issues around existing users, allowing LL to retain them and their loyalty while also avoiding initial issues of growing a new user base.
However, offering such a path might itself create issues. One of the biggest potential benefits in an “SL 2.0” would be the ability to incorporate the infamous “avatar 2.0”, which has been the subject of speculation on-and-off since around mid-2007. This is something that is unlikely to happen within Second Life as it is because of a myriad of dependencies means a dramatic overhaul of the avatar could break things. As such, developing a new avatar form for “SL 2.0” could end up breaking compatibility with “SL 1.0” and render migration either problematic or (worse case) pointless.
Perhaps the biggest issue with any “SL 2.0” though, is not technical, but physical (so to speak). At the end of the day – and as Qarl comments in a recent Metareality podcast – a lot of issues relating to SL are actually centred on the relationship between users and Linden Lab itself. These take a variety of forms, some are justified (such as people feeling the company could be more forthcoming within consistent and more open communications and dialogue with the user base), others are completely unjustified (such as claims that LL are out to “kill” aspects of Second Life or that they act “maliciously” towards users).
Regardless of how justified or otherwise claims and arguments about LL are, the fact is that whatever the platform LL provides, the issues and arguments will likely continue. As such, there is a risk that any “new” SL could be taken to be “same s***, different shovel” by both sides of the relationship; users will continue to bemoan LL and LL will continue to feel they are in an uphill battle facing the same criticisms and complaints they face at the moment. This in turn could lead to both sides asking the question, “Why even bother?”
Over all of this, however, lies the biggest question of all: what, exactly, would LL achieve by taking such a route? It’s unlikely that “SL 2.0” would achieve any grater success than the current Second Life has achieved or has the potential to achieve, allowing for all the new capabilities being developed. Thus, any new variant of the platform is liable to end up occupying precisely the same niche as the current product, with more-or-less the same attractiveness to users and possibly the same grumbles and gripes – and this renders any idea of an SL 2.0 developed by LL pretty much moot. Far better that they focus efforts on improving and enhancing the current platform and in maintaining / increasing its relevancy.
Nevertheless, the idea is still an interesting discussion-point – well, for me, at least!
Inara Pey: Living in a Modemworld 
One problem is that some of us have spent hundreds if not thousands of pounds (or dollars or euros or whatever) in real money inworld. To suddenly be starting as a n00b on a new platform with none of your skins, hairs, clothes, and all would be a massive, massive wrench. To the extent that some would simply walk away rather than spending the money.
LikeLike
Exactly – as I’ve commented in the piece; migration would seem to be the wisest course of action were such a ting to happen, but brings with it a possible new raft of problems.
LikeLike
Oh, without a doubt. It’s a two-edged sword, for sure.
LikeLike
I currently pay for 4 sims for my store — starting from scratch, I would need only a fraction of that. Same with Marketplace sales — would not be much commission coming from that.
As a parallel universe, tho, maybe it would work. (Tho I barely have time now to run one store in one VW.)
However, now that UGC is being redefined to be User Uploaded Content, it may be moot.
LikeLike
Wow, a lot of good points there. To my mind, there will be an SL 2.0, regardless of whether it is LL or a competitor. VWs will continue to grow and evolve. But i think Qarl/you have hit the nail on the head about the Lab/User interaction. There will always be a group offering only destructive criticism and complaining/conflating every fault — a single missed teleport becomes a day without being able to — but on the other hand, LL shows little sign of learning to communicate with the user base or even boast about their triumphs.
LikeLike
Thanks :).
Ob the borader front, I do agree – there will be an “SL 2.0” popping-up somewhere at some point. I just don’t see it as coming from LL (at least as a completely “new” offering), or even necessarily utilising what we see as “grid software” as we use it today…
LikeLike
Well, when discussing SL 1.0 vs. 2.0 there is this nagging feeling of revolution — getting rid of the “old” issues and start with a new blank slate where none of the issues are present.
In theory, that’s what pretty much every self-proclaimed “SL killer” VW has proposed in the past. Where are all of these “SL killers”? Can you still remember the name of the last one? (Aye, it was Blue Mars… or was the last one realXtend? I forgot 🙂 )
If we look around, the truth is that there is really just one social virtual world with user-generated content with visual contiguity (that means: one single landscape, as opposed to separate “rooms” which have no connection between them, except, well, for links — IMVU, for instance, which has grown to be bigger than SL in terms of users, is of the latter kind: a social virtual world with a lot of user-generated content but without visual contiguity). Well, two, because OpenSimulator could arguably be claimed to be a “competing” technology, even though I tend to bundle them both in the same package.
Should that surprise us so much? Well, not really. There is just “one” World-Wide Web — in the sense that a single concept was implemented into a single protocol, which works in the same way, even though there are dozens of Web browsers, dozens of Web servers, thousands of frameworks, millions of applications, and billions of different pages. The concept, however, has remained surprisingly unchanged since 1990.
It certainly has evolved. Here remains the difference: there were certainly a lot of competing hypertext technologies around 1990, and many more developed afterwards. One survived to create the Web as we know it today. But we cannot hardly say that it’s the same Web that we had in 1990. To be more precise, the Web is a combination of the HTTP protocol (to transfer content among servers and clients) and the HTML page specification (to display that same content on the client). HTTP went from 0.9 to 1.0 and to 1.1 in a bit over two decades. HTML evolved a bit more and we’re currently at HTML5. Nevertheless, HTML5-compatible browsers (pretty much everything out there at the latest versions) can still view HTML “1.0” pages (using the 1993 specifications).
You can make similar arguments for what we know today as “Internet E-mail”, which is… 30 years old. That protocol for mail communications over the Internet was firstly developed in 1982. We still send email using that very same protocol, although there were developed hundreds of competing (and often much better implemented!) protocols.
And even today, people log in to remote servers using telnet, which is a very crude (and insecure!) form of text-based remote connection to servers (but Windows, Mac and Linux still include client software for it, and Linuxes still include server software for that — not recommended, as it really uses a very insecure way of transmitting data). It was created… over 40 years ago.
Second Life, by contrast, is a little over a decade old (aye, we’re celebrating SL9B, but that’s for the day it was opened to the public). There is content in the mainland (for example, the Governor’s Mansion in Clementina) which was developed in 2002. You can still view it perfectly with a SL Viewer developed in 2012.
What is my point here? One thing is addressing stagnation, i.e. a technology that never got further refined, and, as such, seems obsolete because of being “stopped in time”. These tend to be discarded as soon as something “better” comes along. But when something is stable enough to provide adequate service, and survives the “test of time”, it doesn’t automatically follow that it gets discarded for something shiny, bright, new, and utterly incompatible with previous versions. Instead, it gets evolved: new things/features are added without breaking the whole concept.
It’s undeniable that Linden Lab has done that with Second Life for the past decade. They have consistently addressed a lot of bottlenecks and issues here and there, and introduced new functionality while still keeping Second Life, as a whole, a functional platform which is still based on the original framework. And, of course, it still keeps the same users. Accounts created in late 2002 or early 2003 are still around today and are active members of the community (I’m just going to name Washu Zebrastripe and Damien Fate as good examples, but fortunately there are hundreds more 🙂 ).
OpenSimulator, to an extent, has shown that it’s possible to tweak Second Life further without breaking everything. Direct megaprim support was present in the earliest versions of OpenSimulator (nowadays, LL updated the size limits from 10x10x10 to 64x64x64 without breaking anything). Things like megaregions, 45K prims, unlimited avatars, regions-on-demand (i.e. the ability to break regions in several pieces and distribute them across several CPUs to allow servers to keep up with the stress of dealing with hundreds or thousands of avatars), saving Windlight settings per region, tons of new LSL functions and so forth show that there is a lot that can be crammed into a “Second Life server clone” without breaking either the client or the “SL experience”; if LL doesn’t actually adopt those changes it’s because of their policies, not their lack of knowledge or some technology limitations. TPVs add tons of support for all kinds of features that LL is not comfortable with. Even the idea that a “new SL” would allow better support on non-desktop computers because it wouldn’t be so tied to LL’s renderer is not really true; viewers like Radegast show that it’s technically possible to create a full 3D experience with SL content displayed on a renderer that doesn’t share a single line of code with LL’s own renderer (as opposed to almost all other SL viewers, which are just a fork/branch of LL’s original code), and still make it very lightweight (Radegast works flawlessly with good FPS even under emulation mode on a Mac!). And by slowly moving away from the “original” data communication protocols (mostly based on UDP and centred on simulator-to-client communication) towards Web-based communications (e.g. HTTP textures, HTTP directory) LL’s engineers managed to push traffic-intensive requests into Amazon’s cloud computing services, where it also can be efficiently cached and distributed much better. So there is really a lot that is going “under the hood” which LL (and, to a degree, the OpenSim core developers) are constantly improving.
Maybe the next “big change” will come as we move from a prim-based economy to a polygon-based economy (better known as “Land Impact” 🙂 ). The amazing thing about that is that LL is willing to change the whole way we think about 3D content hosting (in terms of what metrics are important to charge people for having their content publicly available) while still remaining within the same platform. Remember Blue Mars? Their model was based on the assumption that a “region” could have as many polygons as you wished, so long as you’d be able to fit it all under 500 MBytes of data, and you’d be charged for that. It was viewed as a far better way to charge people for their displayed content. Well, LL eventually figured out a more abstract model for charging tier than using prims, but without needing us to move away towards another platform.
I also agree that the main issue in using a revolutionary model as opposed to an evolutionary one is migration — of contents, of accounts, of a whole experience. I’m not so sure that this can be accomplished easily — a lot of people (and we’re talking about millions of users here!) would simply not “migrate” if it were voluntary, and if it isn’t, the question is if it would be worth all the trouble. After all, if the result is something so much different from SL, why bother?… One could “migrate” to OpenSim instead — which is far worse (in terms of stability and performance) but way cheaper.
Instead, what makes more sense is to continue to increase the trend, launched by Rod Humble, to add more and more features, fix more and more bottlenecks, make some aspects of SL more worthwhile, while still sticking to the “same” platform — just making it evolve. The end result might be as different from SL in 2002 as the Web in 2012 is different from the Web in 1992.
But it still be the “same” technology, the “same” platform, the “same” virtual world.
LikeLike
It seems to me that it would be valuable to be able to look back at past decisions in the light of experience and change them without being concerned with rocking the boat too much. For example:
Was direct teleport a good idea, or were telehubs better?
Is the pricing structure for land right?
Is Marketplace structured right?
There are literally hundreds of decisions made along the way that in hindsight might have been done differently.
Does LL need to make fundamental changes to adapt to emerging technology like tablets and wireless connections?
The most important thing to maintain as much compatibility with as possible is inventory! It is our inventories that keep us in SL as much as the community.
I believe LL should be looking at these thing on a more or less continuous basis and be ready to move when the time is right.
LikeLike
It is a logical decision if there is no possiblity of building on the existing system. I’m sorry to say this but I think a single viewer would be preferrable to the mess that’s been created with so many viewers. The big problem would be the loss of all the incredible content that exists on our SL Grid. IF they could port existing regions to a new SL 2.0 Grid then this is a viable future.
Here’s a lesson I learned when studying successful technology companies, the best of them built bridges from the present to the future. For example, programming langauage that looked extinct have survived because there was an assessment that the talent pool could be trained to use enhancements far more successfully than forcing them to learn new languages. It has resulted in some ugly but functional programming.
I’m in the camp that believes LL’s board has a bad track record on many initiatives. I’d have to hope for the best on this one.
LikeLike
I’m 100% in agreement where past board decisions are concerned, although I’m still optimistic about the current overall direction the company is taking vis-a-vis the new capabilities, etc., albeit it with concern over the company’s increasing withdrawal from direct communications & their seemingly mixed attitude as to whether they are a service / platform provider or a service/platform/content provider (personally, I’d prefer the former and have done with it).
Whether the new direction vis-a-vis capabilities (and on a broader front, the development of new non-SL products) is something that has been initiated by the board directly, or something Rod Humble brought to the table and got the board to support is an interesting question in itself. But for the time being, I’m glad that the company is looking to increase the attractiveness of the platform in terms of a diverse content creation capability and looking to strengthen its longer-term future.
There’s still the thorny issue of retaining incoming users and getting them actively engaged in the platform – and this is something that need to be tackled sooner rather than later in some respects. Initiatives were promised back during SLCC 2011. So far as I can see, none of them have materialised in any real shape or form that has had a real impact on overall retention.
LikeLike
Hi Inara… I understand your preference for LL to just stick with service/ platform provider. It would clarify roles substantially and I suspect they’d be happy sitting back and collecting the royalties. But SL’s potential is hindered by SL’s lack of involvement has made most of the mainlands and eyesore, a hodge-podge of earnest attempts to create beauty offset by mindless junk or flagrant commericalism. A few simple covenants could have made these projects something of great value, like the successful LL projects.
LL has created some very valuable content. Now that the threaded region crossings is implemented and working beautifully, the Lab’s Blake Sea project is a great success and a gift to everyone who loves the sea. Also, Bay City is a great success. The Lab knows how to do things right and i’d even say that Amazon Adventure is a big success (if you replace those rediculous animals and trampolines) for new users. It’s a wonderful place for new users to get acquainted as opposed to those welcome area ghettos.
Concerning SLCC 2011 promoises, I have a feeling that many problems were fixed although I can’t recall all that was promised. I don’t recall if fixing the region crossings problem was one of the promises, but it has been fixed and seems to be working well. Concerning the new user experience, I’m pretty disappointed as well, but like some of the new user attractions like Realms and Adventure. But I’m with you on how pathetic user retention is. I believe the V2/3 vs 3PV with V1 interfaces is a disaster is still adversely impacting that whole new user experience. Simply stated, how can Phoenix user help new users with V3 interfaces? If I bring friends to SL, do I tell them to go ahead and use V3 knowing that i’ll only be able to give them vague support. OOPS, here I go ranting, time to bail. Hugs
LikeLike
Oh, technically, they’ve delivered, I’m not taking that away from them.
The elements I’m referring to are various comments made concerning LL addressing issues of capturing the incoming sign-ups and translating more of them into active participants in the economy and SL. None of those ideas or approaches seems to have any impact whatsoever. Most of these were given via the community-based presentations by LL staffers.
Oddly, where the Viewer is concerned, I don’t see either interface being an issue for new users when it comes to usability. True, many established users see the later iterations of the Viewer (V2, V3, V3.2) as anathema and something to be decried (and is nothing new – users have railed against virtually every single change in the interface that LL has ever made, going all the way back to the likes of v 1.14 and 1.16 and the “radical” changes that came with the likes of 1.18 and 1.22). However, where new users are concerned, the personal bias of established users is actually irrelevant, because it doesn’t matter which particular flavour of Viewer is on offer, the learning curve is still there, and in the eyes of the novice users, no one Viewer is going to appear any “easier” to learn than any other flavour. Bias only enters the equation when it comes down to whether or not the novice user is being assisted by an established users, and where that established user’s preference lay.
The fact is that if there is something sufficiently compelling about Second Life to keep a new user engaged, they will wrestle for mastery over whichever Viewer is offered to them in order to experience what is on offer, regardless as to whether they have assistance or not. So for me, really, the issue is making sure new users can get to the things that attracted them to SL as quickly as possible, and with sufficient assistance in order to get them comfortably started. How this is achieved is questionable – but more needs to be done, and it is the one area where I feel should be collaboratively approached by LL and users as a joint effort, regardless of how well (or otherwise) past efforts faired or were treated.
As to Mainland, having LL set back as a platform provider doesn’t actually limit them in taking actions to work more constructively to regenerate the Mainland and develop projects that could equal the popularity of places like Blakes Sea or the tidiness of Bay City. All it takes is the right approach to the Mainland community, the provisioning of the resources for them to take a more pro-active role in Mainland “management” – if handled correctly, this could even be for the betterment of LL’s position as platform provider.
LikeLike
Thanks Inara… you have great insights and do a great job. And back to the original idea, I just don’t see how SL 2.0 would help anyone who has invested in the SL we’ve created. I’m with Gwen on Evolution and hopefully SL’s internals and architecture have the “bones” to build the future on. And hopefully the LL Board will have the desire to move that platform forward. I’m with you on Rod Humble, he gets high marks with me.
LikeLike
Yup, evolution always wins-out over revolution.
And thanks for the feedback :).
LikeLike