Wednesday, January 31, 2007

First Look into a Mirror

[Update added to the end]
The new First Look Client has a new feature that the recent blog post didn't say anything about. That feature is Dynamic Reflection, and that means working Mirrors in SL. I haven't taken a look myself yet, but Gwyneth Llewelyn has made a video and put it on youtube.

This looks pretty cool if you ask me. Now this does not mean this will be in the normal client any time soon. There have been features in Beta before that never ended up in the client, because they didn't work(well) for many people. But this is cool non the less.


In response to Gwyneth's vid, ZATZAi made a longer video showing how the new mirror feature works on different shapes an shiny values. Even Mirror to mirror reflections work.

Not only shiny prims mirror but also ripple water reflects!
Picture by Chandra Page

And i love Chandra's comment on it:
"It's like shiny, but it's shiny on fire, being chased by ninjas and helicopters, and the ninjas are on fire, too."

Tuesday, January 30, 2007

Swift action on Land malaise

The price of land in SL has been rising and rising, because LL had a huge demand in island sims and couldn't add sims fast enough to the mainland continents. The last 2 weeks they started to release more mainland sims and those went in auction for over 3000 us dollar. To meet the higher demand Robin Linden annouched today that they are going to open a new continent to the east on the map to meet demand.

Robin: "Over the last several weeks we have been adding land to the mainland at a fairly slow rate (4 new regions per day) while we complete the delivery of islands ordered at the end of 2006. The result is that pricing has been creeping up steadily and is currently about double what we saw per m2 in November. In order to meet the increased demand, our plan is to start to increase the rate at which we put mainland regions online.

Look for a substantial new continent to begin to appear off to the east!"

This is interesting because there are a lot of new users who have only seen the land price rise and might be caught by surprise of sudden Linden action that makes the land prices drop. We might see some new land barons make some huge loses.

Also I hope they make some effort to finish the borders in the southern continent and the old core mainland. There are still some spots left that just abruptly end, if I'm not mistaken.

But Robin said something even more interesting.
Robin: "This increase in land supply should also help to offset the impact of the use of ‘bots as high speed land scanners to skim off the lowest priced, and generally smaller, less desirable parcels. In addition, we’re looking at adding a step to the purchase process which should make it more difficult to use ‘ bots in purchasing land."
If you haven't heard of it before. I think as of last week a landbot is active in SL. A completely automated program based on LibSl that scans the land for sale listing faster then a human can and buys any land that is under market value, to put it up for resale again for a bit higher price offcourse. While I don't see how increase in land supply will really help against landbots, it is interesting to see that LL is already looking in to ways to prevent such bots from operating. It will be a amusing arms race to see.

Who's first?

A thing what is still coming up is the first in SL claim. A while back I talked to a friend here in the Netherlands who had talked with some museum about doing something in SL. One of the first things they said, "Oooo, we can send a press release that we are the first, ... " *sigh*
It annoys me to no end.

Yesterday I was talking with Adam Broitman of A Media and asked if 'something'(have to be vague here) would be a first. While the exact thing might be a first, there is a lot already that would force the first claim to be pretty narrow, IMHO, and be at risk to be ruffle some feathers.

Eric Rice posted today about first-ism in the blogosphere and made some good points. I pointed Adam towards who blogged a response to it.

Eric ended with:
"If you give me a choice between being first or being best, and I'’ll strive for being best."
And Adam's opinion:
"My bottom line; go ahead, be first. Be first at everything you do if that is what your gut tells you, but be calculated. Make sure it is the right move. Don'’t be first for the sake of being first."
My own opinion is that,I don't mind if something is first. But you have to be aware where you are the first, because if you enter SL first, it only means you are the first on the track. You can shout and waste energy about it, but the triathlon still has to be run.

It will just give the impression that you care more about being at the start and getting some publicity then actually performing well. You can better use that energy to tell why you are doing it and how it will help you and/or your clients. People will know the difference and greater credit for being a leader in your field will follow in due time. If you actually where the first on the track will not matter in the end if you didn't use that extra time figuring out why you are here.

Sunday, January 14, 2007

26k laggy souls.

26142Today we broke the 25k and the 26k barrier of peak concurrency in SL.

And as you can read here, it caused some troubles. I Guess i was overly optimistic last time, when i said high concurrency wasn't really causing problems. Well, it is not the only thing that is causing it. :P

26k is a 30 procent growth in 16 days, when SL broke the 20k barrier. If any statistic should matter anything then it is this one, and that is a pretty good growth. Seems like we can reach 30k before the end of the month, if only the grid+monkeys can keep up with it.

Tuesday, January 09, 2007

Metaverse: Immortality

There has been so much coverage of the recent announcement of the SL client being released as open source that it feels unnecessary to repeat the arguments of why it is so crucial to get all the innovative developers — the ones crying for more features — outside of Linden Lab being "brought into the loop", and leave the project management details to LL's own team, as well as the "boring" issues of stability, scalability, and (minor, but relevant for some) bug fixing.

However, the interesting aspect of having the SL client becoming open source — and very soon, the server code as well — is to make essentially Second Life "immortal".

Let me explain this simple concept. While many people are far more comfortable in having Linden Lab "run the show" (and not the open source community!), they're also afraid of what happens if either an earthquake strikes San Francisco, or Linden Lab going bankrupt, or some other catastrophe occurs (natural or financial). Well, to address the redundancy of the grid, Linden Lab has now servers on a co-location facility in Texas as well. But that is still not enough.

Now, with having the source code available, we can have a peaceful sleep. If the competition (the one that gets mentioned every time but failed so far to make an appearance) crushes Linden Lab out of business, or if Philip decides to move to an island in the Caribbean and forget all about the "Metaverse thingy" — we, the 2.5 million users of Second Life, can go on.

Because we have the code.

One would argue that it's impossible to set up a volunteer-run grid with 5,000 servers overnight, even with the code. I agree. It won't be the "same" grid — ever. It will have shaky servers, some run from universities, some allowing only 3-4 avatars on a sim, and being run from people's home ADSL connections. But — it will be a grid. Of sorts. And it can only become better and better.

Imagine the entities having serious stakes in Second Life now: NMC, IBM, the Metaverse Development Companies. What would they do now if SL disappeared from the face of the earth? They would shrug, grumble, complain, but move to other things instead. And SL would be quickly forgotten like so many other attempts before it (like Lombardi's ViOS).
But now things have completely changed. They could inhale deeply, shake their collective heads, and say: "ok, LL is out. But we'll keep the torch burning. Let's deploy our own grid and use LL's source code to develop it further". The Big Corps would start rolling out their own sims immediately, since they have the resources to do so. And we'd just get some notice on the MOTD: "Linden Lab is shutting down their grid as of [a date in the future]. But you can connect to [insert Big Corp or Big University name here] grid instead, just point your SL viewer to their grid instead. Click here to know how to do it"

[Second] Life will go on :)

Thus, no matter what ultimately happens to Linden Lab's own grid, the Metaverse built upon Second Life has at least reached immortality. It has critical mass — 2.5 million accounts is, no matter how much we can discuss it to exhaustion, far better than "nothing"! — and it has now the means to "go on beyond Linden Lab". This is quite a difference from everything we have been seeing as "competition" in the recent past — all of them are based on the model that you have to trust the company to stay around. Linden Lab effectively moved the trust into the hands of the users and the product.

Old "games" and much software who went open source at some point lived way beyond the "corporate life" of the company producing it. Of course its growth might have been more reduced — without active promotion from a company, it's hard to make an "old" project grow (which explains why MySQL, although being an inferior product to PostgreSQL, is able to grow much more in world-wide attention: it has a company behind it promoting it actively).

Again, Second Life has now an advantage: it has engaged companies and universities to co-promote it. So at the end of the day, it will be the megacorps promoting the SL metaverse more and more, and Linden Lab less and less. We're not there yet, but we'll be — we can, these days, look at the source of press releases and articles about SL and see this happening. In 2003, the only company promoting SL was, obviously, Linden Lab. Nowadays, how many PRs from LL do you see every day, compared to articles from all other sources? So I think we're coming very near to have "critical mass" for promotion.

In conclusion: we have all the requirements (well, missing the open source server; that will come "sooner than later", and now that we know Linden Lab keeps their promises on "going open source", they'll keep that one as well). We have the know-how. We have the strong user base. We have the content creators. We have the support of the megacorps. We have the vision, skill, and talent from the universities.

And we have the source code.

Armed with all that, Second Life will not die a premature death.

As a final thought, one could argue that things like OpenCroquet would be able to do the same (and, in some areas at least, OpenCroquet is "better"). But the ones betting on OpenCroquet or any similar tool are totally missing the point. Remember what it takes to make an open source project thrive — it needs support from everybody involved in the value chain. OpenCroquet, like many similar products (VR and not VR), has strong academic support. It has been the source for more PhDs and Master degrees than any other VR platform. But... the user base is tiny, compared to SL. The non-userbase support comes from academia, not corporate business. It's also way harder to use. There is no economy. There is no incentive to use it, except if you're a top programmer without time restrictions or work in an university trying to do a masters' degree on OpenCroquet. There is no grid. In essence, there is no Metaverse around OpenCroquet — just users and programmers and academic support.

We can now understand a bit better what the "Metaverse" is supposed to be. Not the "ultimate VR tool" (as some friends of mine are fond of saying: "I'll use SL if it gets photorealistic texturing and far better images"). Not the "tinkerer's paradise". Not the delight of happy sysadmins that love complex networked systems. Not at all. The Metaverse will include all those — and they are also crucial! — but they will add the content producers. The Big Corps who want virtual presences. The small SL businesses offering new services. The middlemen (and -women!) who add value on top of programming and content. A stable economy. The global community. The entertainers and the consumers of entertainment. All of them are required.

Second Life, as we know, is the only tool that has all of the above. Others might have better graphics, a more solid business model, support from either academia or industry, or excellent programmers, or even great content producers. They might even have a better model for their grids — technical and financial. But they all lack a user base — they are "brand new things" — a society, an economy, a model of developing relationships. How long will it take to have all the above emerge from a "new thing"? 4 years at least? Well... you see my point.

Nobody in their right mind would now start things from scratch. Even if you want to get into the "Metaverse Thingy", be clever: don't reinvent the wheel. Grab the source code and develop your own client; then wait for the server source code to be freely available, and tweak it. Then offer services on top of that infrastructure — services that nobody else wants to offer, not even Linden Lab. Who knows, we might even have the "IBM Metaverse" quickly enough (in a year or so).

But the beauty of it is that it will be SL-compatible :) And that is why I think that Second Life has achieved immortality.

Tuesday, January 02, 2007

Second Life and the Gartner Hype Cyle

syndicated from _notizen

Critics and fan bois of Second Life seem to discover the "Gartner Hype Cycle" all of a sudden. Funny, since the idea of hype cycles is one of my all time favorite memes. This model describes how a new technology "appears in the world at large"; how they are mostly unknown at first, then suddenly public interest explodes, implodes after the press looses interest and finally reaches a kind of "normal" level. More on that here.

It is perfectly normal for me, to look at any new technology, like Second Life, in terms of a Hype Cycle. But the idea of a Hype Cycle just models public attention of "something new". You can not use it to debunk the "something new ("Baah ... it is just a hype"). Nor does it make sense to try and baptize it with hype cycle terminology ("Just you wait ... until the plateau of enlightenment.") What most, who let this term elegantly drop into a discussion, like to forget is, that a hype cycle doesn't care about the usefulness of a technology, doesn't care about the acceptance, the number of users, doesn't care about the volume of business ... it only models the volume of buzz in the press.


Most people using the term currently see Second Life at a certain point in this curve and draw conclusions from this assumption. The conclusions (and assumptions) are colored deeply by the authors own attitudes, though. Those who are skeptical, see Second Life already on the downturn, following the Peak of Inflated Expectations. Even some usually level-headed commentators tend to confuse a little critisism with a peak that has been past. Some of the biggest fans on the other hand like to locate Second Life already at the Slope of Enlightenment or even on the Pleateau of Productivity. Both attitudes are ... highly debatable, to phrase it politely. Please recheck Gartners own definition of Hype Cycles!

Of course there are inflated expectations about Second Life - and of course there are critics and skeptics who are crying "it's all BS". A backlash is starting. There are critics with reasonable arguments and some of the "i don't like it, therefore it's doomed" school, who, while reflecting personal preferences, might have a large following. A huge problem is the coverage in the "professional" media, which is not very thorough and results in some misunderstandings with those critics, who rarely have the time for collecting personal experiences. Often it is not the platform "as is" which is critisized but assumptions, promises and euphoric fan reviews. This is perfectly normal with a Hype Cycle, too, and doesn't say much about the the technology being hyped.

Hype Cycles have a much longer wavelength than most critics and fan bois would like to see. The peak of media attention with Second Life hasn't been reached yet - nor have we seen the peak in public criticism. Maybe both will happen in 2007. And the "Slope of Enlightenment" or the "Plateau of Productivity" is at least one or two years away. (Which does not mean that no one can be productive in Second Life now; please check Gartners Definition.)

But no matter where you see Second Life on that beautiful curve: it is impossible to extract any meaningful assessment out of that, if Second Life (or the idea of virtual 3D worlds, of which SL is just an early and imperfect implementation) will prove to be a useful and widely accepted technology in the long run.

I believe, that virtual worlds like Second Life represent a development, that will prove itself to be (very) useful. And I can see indications for this in some of the projects already being done in Second Life. But it is impossible to prove - now. As it is impossible to falsify this assumption. Time will tell. Terms like "Hype" oder "Hype Cycle" are extremely useful for describing the workings of public attention; and completely useless, when trying to judge a technology.

Monday, January 01, 2007

Concurrency of 21115


It is only 3 days ago that we broke 20k and now we are on 21k+.
And to be fair, while we have all been complaining these last few weeks that sl was slow whenever concurrency was high. Lately SL seems to be behaving much better under high user loads. Keep up the good work LL.

Breaking records is allways a good way to start a year. :D