The Emerald Implosion

Monday, August 23, 2010 Monday, August 23, 2010

If you’ve been reading any SL blogs lately, you’ve probably come across the incredible events surrounding the famous (or infamous) Emerald Viewer. Resignations, reformations, scandals, possible criminal acts and other mayhem ensued throughout the week. I’m not going to detail any of that stuff; it’s been well-covered elsewhere.

Instead, I am going to bring forward my thoughts on how it came to this. There’s definitely reasons why this week’s implosion occurred, and why it may not happen again. All these events are connected at a very high level.

But let’s start at the beginning.

We have a certain company, Linden Lab, who market a very unusual product: Second Life. This product is amazing, but it is also a very complex thing to deal with. In fact, it’s so complex that no one really knows how it should be set up.

The product, SL, is so complex that while it is amazing, relatively few people from the public are able to manage to successfully use it and stay using it. Everyone uses the standard viewer through which everyone experiences the product.

Advanced SL residents grow to want more from the viewer, as they’ve managed to learn many things about the environment, well beyond the basics. But they’re frustrated because they have only one option.

The Lab recognizes this need and responds by open sourcing the viewer code. They hope that the community will adopt the freely available viewer code to develop the advanced features that it wants, while leaving the Lab to put their limited resources against other problems and ventures.

The Lab focuses on growth. They believe that to attract more residents, they need to somehow simplify the experience so that it doesn’t scare people away. Indeed, the survival rate for new signups is abysmally low, perhaps as low as 1%. One of their simplification strategies is a less complex viewer. They begin a project to develop this new simplified viewer.

Several groups adopt the open sourced viewer code and begin tinkering. Some happen to be professional or near-professional developers, but others are not. Hackers and griefers also take a stab at making their own viewers - sometimes for nefarious purposes. A variety of viewer options emerge, all with differing features, support, release schedules and reliability. Some residents try them and begin to have opinions on their favorites, usually based on their particular needs.

One third party viewer (TPV), Emerald, becomes somewhat more popular than others, perhaps based on its frequent release of interesting and unusual features. This viewer is in fact the opposite of the Lab’s work: it’s a complex viewer including *more* features, not fewer. But these features are well-received by many long time residents in the community.

With popularity, more information comes to the surface about Emerald and the folks behind it. It turns out that several of them have known histories as griefers, some being suspended from SL in the past. It is further discovered that mysterious encrypted information is being sent from the viewer to Emerald’s server. The Emerald team does not reveal their real identities, thus making it very difficult to ascertain their level of responsibility.

Aside: it was at this point I concluded it was too risky (at least for me) to continue to use Emerald. Code written by anonymous former griefers, known to be sending unknown information to parts unknown, was simply too suspicious. I, and several others, deleted Emerald from our systems and changed our passwords in case they had been somehow recorded by Emerald. I feared an incident of some kind would occur at some point in the future and didn’t want to be part of it.

The Lab releases their new, simplified viewer: Viewer 2.0. Amidst fanfare, V2.0 included features intended to simplify things for new residents, but for existing residents it was too different, too simple and worse, beset with annoying bugs.

Viewer 2.0 becomes the default viewer - but because it doesn’t match resident’s needs, they flock to alternatives. Which one should they choose? Emerald was the most popular of the TPVs, and it’s usage grew significantly. Legitimate developers join the Emerald team, and it continued to be improved with additional features. Emerald gained many supporters as residents tune into its unique features.

Suddenly, there’s an incident.
The Emerald home screen was modified by one of its developers to perform an attack on a rival site, thus using the computers of all Emerald users for this activity. Poor judgement? Yes, indeed! Just as I had lost trust in Emerald months earlier, this incident resulted in a loss of trust by many former Emerald supporters. In fact, Linden Lab removed Emerald from its official list of TPVs.

The Emerald team breaks apart due to the incident and its aftermath, but reforms under new, hopefully more professional management. Time will tell if this is so, as trust is easy to lose but very hard to gain. Good luck to the new team!

But both problems still remain: existing residents need an advanced viewer and new residents need a basic, simplified viewer. Neither group is adequately served today, and Linden Lab needs to develop a strategy to address this critical issue before they will begin growing again.

SLCC 2010: The Resurrection

Thursday, August 19, 2010 Thursday, August 19, 2010

It was with great nervousness did I watch the unfolding of 2010’s Second Life Community Convention earlier this year. I had attended two previous instances of this famous event, 2008 in Tampa and 2009 in San Francisco. The 2009 event was announced rather late in the cycle, leaving little time for presenters and attendees to get organized. It nevertheless took place, fun was had by attendees, but the convention’s sessions and organization were not exactly optimal.

Meanwhile, this year’s event almost never happened.

Weeks passed, with no announcements of any kind regarding the 2010 event. Like those who attended previous events, I was looking forward to attending and meeting up with virtual friends. But no announcements. Was there even going to be an event at all?

By March I had personally concluded that it must be over. There is no way a major convention can be organized in such a short time, particularly by SLCC’s previous organizational track record. Saddened, I wondered how this would play out. How would the SL community go forward without a foundational event around which we can come together?

Worse news: Linden Lab suddenly revealed massive staff layoffs and a vague change in direction. The Linden Dollar significantly slipped in value for the first time in eons. Both the annual convention and Linden Lab itself appeared vulnerable and perhaps even in jeopardy.

Amazingly, SLCC 2010’s date and location were announced only weeks before the actual event. This is incomprehensible as far as event planning goes; there is no way you can expect to have a successful convention with only a few weeks notice. Nevertheless, to support the community I registered immediately, not knowing what would happen.

I arrived in Boston a day early and learned the dramatic sequence of events: The previous SLCC’s were run by “The Future United”, a group which earlier this year folded when the last board member admitted they could not put on the convention. It was at that moment that concerned and leading residents put together a plan to save SLCC: the new body, AvaCon, formed and then began lengthy negotiations with the Lindens to secure the licenses to put on the official convention. Only after these documents were signed could AvaCon announce dates and location - and then they had only eight weeks to put it all together for real.

And they did, truly magnificently. I’ve attended many conferences and this one was run as well as any. The sessions were interesting and well planned; the hotel was near-perfect for this event (especially the blazing fast and free WiFi network); the off-hours events were terrific; the logistics, program, badges were better than any previous SLCC I’d attended. I give my highest compliments to the New SLCC organizing team, led by she-of-no-rest uberorganizer Fleep Tuque, for doing an impossible job in an impossible time.

As for SLCC 2010 itself, I remember it as a series of moments:

Gwampu Lomu singing with the totally awesome Tamra Sands. Well, at least he sung the last note of the song, anyway. Tamra is a far better singer than Gwampa.

Listening to the endless puns from master comedian Lauren Weyland, whose distinctive voice could be heard almost everywhere. Example heard in the extraordinarily elaborate main ballroom: “This room must have been built when America was China!” Have you found your shoes yet, Lauren?

Having an impromptu iPhone FaceTime party in the executive lounge, where several of us FaceTime virgins realized we could video call each other! So we did across a coffee table - a true NerdOut.

The discussions and arguments taking place every evening in the lobby, bar, restaurants or suites that ranged over any conceivable SL-related issue. For certain, we all had the answers to every problem, or so it seemed.

Meeting the notorious Prokofy Neva in person, who is actually more reasonable than many blog readers may think.

Spending time with the IBM virtual reality team, including the brilliant Zha Ewry, the delightful Dale Innis, Oura Scribe and the lovely Ahuva Heliosense, whose incredibly sultry outfit was seemingly ruined by the absence of a single rhinestone on her rather complex shoes. Or so she thought. But I didn’t.

Meeting a most amazing person: Pooky Amsterdam, whose dramatic presence simply took over every room she entered. If you’ve heard her voice on one of her shows, let me say that her personage matches her distinctive voice in every way.

Taking over the famous Cheers bar in Boston, with perhaps thirty avatars occupying the entire section. Fortunately, Cheers did not know our real names.

Meeting the SL-banned Woodbury group, who surprised everyone by showing up to register for the conference. After initial fears of RL griefing subsided, they were an interesting group to speak with in the dark hours of the night.

Observing once again the astonishing real-life charisma of Philip Linden, who simply bent over backwards to meet and talk with everyone and anyone. Philip spent considerable time talking to the grumbly educators after the surprise announcement of the denouement of the languishing Teen Grid. It was also fascinating to watch certain women simply melt in his presence. One unnamed melting individual kept saying, “I want one of those”, while gazing dreamily at Philip. No, it wasn't Pooky!

Discussing the surprising outcomes of permitting intergrid teleports between SL and Other Grids with Philip, who has evidently thought quite deeply about the subject. I’m sure we’ll hear more about this in the future.

The never-ending post convention “secret” party, which was not so secret because it was being broadcast live over UStream. As soon as the stream went up, missing attendee Eshi Otawara videoed in and joined us virtually. Miss you, Eshi!

Spending time with Noona and Noele, who were perhaps the most unique attendees: they were simply residents! They did not own virtual businesses, perform any artistic works or organize anything. They simply enjoyed SL shopping and decided to come to SLCC for fun. While there were few sessions directed to them, their perspective on things was quite different from the ultra-creative types who normally inhabit the convention.

The continual exhaustion and exhilaration caused by endless hours of meeting new avatars and discussing everything. Some were totally spent, yet could not leave the party - like Dirk McKeenan who fell asleep on the lobby floor at 4AM. Dirk was an unstoppable force. Until he fell asleep.

A windup dinner involving all who assisted the convention at the longest possible restaurant table. I squeezed in beside Filthy Fluno, Prokofy Neva and Fleep Tuque, and had delightful conversations on a wide variety of topics.

Appreciating the several exLindens who still came to the event, even though they were no longer employed by Linden Lab, including Kate exLinden, Joppa exLinden, Teagan exLinden and Pathfinder exLinden. They are those who truly support SL. Yes, that is Joppa stuck on the window. Don't ask.

Watching the full-length documentary “Life 2.0” with many other avatars, and discussing it afterwards. Indeed, I found myself in the middle of a debate on the merits of the piece with Prokofy Neva, Tuna Oddfellow and Zha Ewry, a highly unlikely combination of avatars to be seen. I liked the film immensely, as it accurately portrayed (and made you feel and experience) the confusing emotional transformations that are so common in SL. This film is unlike any other SL-related piece I’ve seen, and if you are in SL, you must see it. How? You will be able to see it next year on the Oprah Winfrey network.

Catching up with old friends (well, they’re not old, the friendship is) such as Feline Slade, Daphne Abernathy, Roland Legrand, Loki Clifton, Ham Rambler, Beyers Sellers, Sloan Skjellerup, Apple & Nasus, Katydid Something, Rhiannon Chatnoir, Tuna & Shava and of course Dirk McKeenan and many others.

Meeting wonderful new friends, such as Explorer Dastardly, Spiral Theas, Anastasios Aurotharius, Landau Shippe, Tamra Sands, Fleep Tuque, Filthy Fluno, Winter Nightfire, Maggie Marat and so many others I cannot remember them all.

But by far the most special moment for me was also a special moment for two others: Chestnut Rau and Zha Ewry, long-time partners in SL who had never before met each other in real life. In fact, Chestnut had not met ANY avatars in real life before this event and was quite nervous about doing so. I am happy to say that I had a tiny role in their meeting.

Fortunately Chestnut had reluctantly posted a rather obscure RL photograph of herself some weeks earlier, and based on that image I was able to identify her standing in the hotel lobby. I approached her and introduced myself, but she seemed a little awkward as those were her first moments meeting avatars. We had a brief discussion, when suddenly over her shoulder I observed someone entering the lobby pulling a suitcase: Zha Ewry.

I immediately held up my hands to beckon Chestnut to stop talking and said in a totally calm voice, “Turn. Around. Right. Now.” She looked at me quizzically and rotated. In less than a picosecond she recognized Zha, who did not notice her. I will never forget the look on Chestnut’s face at that moment, as her emotions radiated brilliantly in all directions. After a moment she strode forward cautiously and began speaking to Zha, who still did not yet understand who this strange woman was talking to him in the lobby. But suddenly he realized it was Chestnut, and the rest I cannot say.

One memory I will definitely not cherish was the loss of my camera, which somehow became separated from me during the convention. I did not notice its absence until I returned home - having used my iPhone for most pictures. If anyone has seen a stray Canon SD980, give me a shout!

There was one more important impression I observed: the announcements from Linden Lab. We’ve all heard the new mantra: Fast, Easy and Fun, which appears simplistic is in fact what’s needed, if it can be pulled off. Philip’s announcements told a story of how they are going to do it by reorganizing their development process and focusing on critical bits first.

For sure, Philip didn’t mention a number of things that people really want fixed (search, for example), but that’s not fixable within this year - search is an extremely complex topic that will take time to sort out. I checked later on with Tiggs Linden, who is in charge of server development. He confirmed that the list of items shown by Philip was indeed achievable, although it will be tough. I’d rather have a shorter list of things that will actually get done than a longer list that won’t. Remember, the list shown was only the 2010 list. There will be more lists in the future, and that’s the nature of the new development process: fast iterations focused on specific features. The impression I got was that the Lab is very serious about fixing these things - and fixing SL overall.

SL is not dead. SLCC is not dead. One was resurrected this August, and the other is underway.

Avoiding The Spiral Of Death

Sunday, August 1, 2010 Sunday, August 01, 2010

Fast, Easy and Fun. How has it come to this?

Fast, Easy and Fun is the new direction of Linden Lab for their product, Second Life. It’s a reaction to the troubles they’ve gotten themselves into in the past year or so. Here’s how I see the sequence of events:

  • A company discovers and builds a service around an amazing concept
  • The concept, while breathtaking, is so new it’s not fully understood by anyone yet, even its makers
  • The service is implemented in the only way possible, given everyone's limited experience with the concept: it’s difficult to use
  • Reasonably large numbers of early adopter users, who are tolerant of difficulties, begin using the new service
  • Beyond the early adopters, few people use the service because its difficult interfaces are forever past their capabilities and tolerance
  • Company notices growth stalling as the pool of early adopters is exhausted 
  • Company seeks ways to break into the mainstream market by simplifying the system
  • The simplification attempt fails because the system is still not fully understood
  • A few early adopters begin to leave for alternative services
  • Management panics and makes big changes. New approach: Fast, Easy & Fun.

When I say “not fully understood”, I mean that while we may fully understand the use of the existing tools and interface, we really don’t understand the *right* way to do the interface. For example, the permissions system is simply unable to deal with many day-to-day situations. How should someone be introduced to the service such that they don’t immediately give up? No one really knows the true answers to these any many other key questions yet. This is a system still figuring itself out. It’s still an experiment.

Experiments are for early adopters.

I’ve written of this before, but it is a phase all companies must endure: the transition from initial “early adopter” mode to a more mainstream “easy to use” mode. Some tech companies don’t survive, others are utterly transformed and some are overtaken by alternatives that have a better approach.

But that doesn’t answer the question of the panic. Why panic so much? Why layoff huge numbers of people? I suspect it has to do with what I call “the Spiral of Death”.

Consider the situation myself and many SL shopkeepers find themselves in: we spend huge amounts of time setting up a shop in SL, making products, doing promotions, etc. and just barely make a living. Suddenly, some users panic and move away to other grids. They say “it’s wonderful, please join us over here!” It may be wonderful if you are, for example, an educator or business person who simply needs to build some type of installation. But if you’re a seller, it’s quite a different situation.

A seller needs a market, and right now SL is still by far the biggest grid with the largest number of potential buyers on it. Yes, there are OpenSim grids all over, but all of them are tiny compared to SL. It’s hard enough to make a living in SL even with its giant user base, let alone grids one one-thousandth the size. It’s even worse because in order to gain any significant amount of OpenSim market, you’d have to set up separate shops in multiple grids (thus spending time building multiple shops and paying multiple tiers. OpenSim grids may be cheaper per square meter compared to SL, but when you need several parcels on several grids, it probably costs more). It just doesn’t make economic sense to move virtual businesses to other grids.

But it would make sense if large numbers of users begin to bail from SL and move to a single popular alternative grid. If that happened, then SL would lose something very critical: its size. When the market moves, so will the sellers. And with them will go even more buyers. And so on, developing into “The Spiral of Death”. Once the user base starts shrinking, it will inevitably go down, faster and faster.

I suspect this is what Linden Lab fears the most. And it’s certainly worth a panic, because they risk their entire business. It could happen: more and more OpenSim grids are appearing, each with some type of innovation or another, but none with massive market share. Yet. Eventually, one of these grids will accidentally or on purpose hit on the right combination of services and interface and they’ll begin to win users over. Linden Lab, if they are to survive, must prevent the Spiral of Death at all costs.

That’s why there’s an “all hands on deck” mode and a focus on the basics. Fast, Easy and Fun are indeed the basic keys to success. If these are achievable, then existing early adopters will stay and newcomers will not abandon the service nearly as much as now. That should result in a growing service once again.

But only if it works. I’m terribly interested to hear how “Fast, Easy and Fun” will be achieved. Perhaps Philip will enlighten us at SLCC 2010?

Related Posts with Thumbnails