DISQUS

Of course not, Two Worlds, or LL's own technology in 2003 would never have a chance to compete in 2009 ;)

Two Worlds, if your point is, "there are more registered users in Facebook", yes, sure, I'm sure of it.

If your point is, Facebook is way easier to use than Second Life, sure, you're right again. Even though there are easier social networking tools out there (Twitter comes to mind ;) ), Facebook is way easier than Second Life... and will be... ever.

It's like saying "IRC is easier than Plurk". Oh yes! But just because IRC is easier to use, it doesn't mean that the number of Plurk users is not growing...

If you mean that none of your friends have embraced Web 3.0, and you feel alone out there, while everybody you know are happily embracing the last remains of innovation on Web 2.0... well, sure. It's always hard to start something from scratch. Guess what, all my contacts on Facebook, without a single exception, are on Second Life! But they're just a small part of out of the 1000+ contacts I've got in SL... not everyone has a Facebook account. Yet :)

What does that "prove"? That both you and I are anomalies, anedoctal evidence, and outside the norm — so you can't base conclusions on either of us. That's why we have specialists for doing metrics, which will carefully trim out the extreme cases which don't really "fit" the data.

As for "SL is rarely fun", I'm sure it is, if your idea of entertainment is: "someone entertain me, or else I'll get bored". SL is for people who think the opposite way, ie. "entertainment is what I do to have fun".

"A failing software company with a rapidly stagnating userbase using a new performance metric in order to insist their product is still relavent? Fascinating. Tell me more."

Sure, the company with the new performance metric is Nielsen, and they aren't either stagnating, failing, or even a software company — they're a market research company ;)

Oh, you meant that LL is a "failing software company".... right. You know, I've heard of three companies that same old story over and over again, it's not so fun any more: Microsoft, Apple, and Linden Lab. They all have "failed to fail" and are still around, in spite of all doomsday predictions.

Yes, it's utterly amazing. I wonder what would happen if LL just advertised a bit...

A slight correction on my numbers: if you add online advertising (which is mostly Google, of course...) to the figures above, Linden Lab will just have a market share of roughly 1% of all digital content sales. But... on the other hand... they would be below the giant Google (and other online advertising giants, including Yahoo, AOL, Microsoft, etc.) but still slightly above Apple!

Facebook or MySpace probably wouldn't even show up...

... but we're soooo few, Cindy! *waves back*

I guess, though, that these days the major focus of attention will become the new new Adult Continent anyway: at least people arriving there by choice, not chance, will know exactly what they're looking for, and they might soon become SL's group of most faithful customers...

Interesting definition, Hypatia... and intriguing how that relates to SL as well. I guess that SL should become "more like Twitter", and that is not so hard to do: the current "conference calls" should be far more easy to do (instead of creating a folder, looking for calling cards, dragging them into a folder, starting the chat...) and could become "lifeblogging threads" (in the sense that they could be "followed" by anyone, if you had them publicly listed, e.g. from your profile), and that and group IMs should both be "loggable" even when you're offline.

Hmm. This is definitely worth thinking more about. I suppose that combining it with the Group Notice tools (which allow embedding of objects, notecards, textures, landmarks... but not URLs!!!!) would make the communication aspect of Second Life look more like, well, Facebook (and not so much like Twitter), but it would be a very, very intriguing solution.

As a matter of fact, in terms of software development, all that LL has to do is simply to grab public, open source tools like laconi.ca for microblogging or BuddyPress for a simple, Facebook-style interface. And it would be nice to be able to, say, publish your in-world "conference call" to a RSS feed — or even directly to Twitter — and get the replies back somehow...

Hmm. Yes, I see there is quite a lot to do on that area that would be quite intriguing to do. Alas, it's a bit out of LL's own core business. But, who knows... people are embedding 3D apps inside Facebook, why can't LL start from a 3D app and put Facebook/Twitter-like communication and information sharing tools inside SL? :)

Well, often the statistics are right, but our interpretation of them isn't :)

I totally agree with you, though. There is no easy measure to value "connectionism" (I like that word!) and how "important" that actually is; and, also, as you so well said, the issue is that few manage to turn "connectionism" into a really sellable product.

To give a (perhaps) stupid example, anyone paying Scoble or Fry to tweet about something, would reach a huge audience. So they have "connectionism value". However, how can you measure that when compared to the number of Oprah followers, who are basically following a 'bot, without any real human being behind those words? (Note that I'm not saying that Oprah shouldn't be on Twitter; rather the contrary: I'm saying that she should be on Twitter, even with less followers, but she should be writing the messages, not a bot....)

Hypatia, that mostly depends if Twitter has indeed a method to communicate their API data to Alexa or not. If they have it (and, as said, I most definitely would create a way, if I were Twitter's owners!), it's actually very simple to do. If they don't have it, they would seem even more naive than I though: a company without a business model, that rejects all offers for buying basically a bunch of users (the technology is nothing special, after all, a lot of clones exist), and that doesn't even worry about correctly capturing their user's data?

... weirdly enough, I use Twitter probably far, far less than the average Twitterer, and I'm #10 on the Second Life Elite of TwitterGrader?!?! Clearly something is very wrong!

Oh, by contrast, you figure out what Twitter's about in one minute or less :)

I tend to completely agree, katemir. I know of very, very few exceptions to the case, and in almost all of those exceptions, these people only came back to SL by chance, because they went to a workshop or conference about SL and learned all the uses that people are giving to it and went "aha, so that's what it's for!!"

But these are really exceptions. The first barrier are the very first 15 minutes, if it loads too slow, or you happen to rezz alone without anybody around, and can't see what's going on, you will give up almost immediately. If you're lucky, meet a few people, and your computer and connection is good enough to navigate through SL easily, you might start chatting a bit, asking people things, and will start to explore one or two areas that might be interesting for you. If they exceed your expectations — ie. they're far more interesting than you thought they would be, you'll stay.

However, well over 90% of all people who log in never reach that point.

Sorry if I sounded condescending about chat history; that was definitely not my intention. In fact, I forgot to mention that to get at past conversations, you need to turn it on explicitly from Preferences. Granted, that won't give you a searchable facility inside the SL client, but only outside SL; and it's also true that if you're not logged in, you won't get access at past conversations that happened on the place you were last logged in (although here is where the replication of a 3D environment works against that kind of thing: if you're not in a place, why should you keep getting messages from that place? ) That's in fact a difference between real-time communication and asynchronous one — like on IRC, if you log back to a channel after a few days, you won't get the past conversations that happened on that channel since you last left it. Of course Facebook and Twitter work differently!

So, yes, SL does not have asynchronous, ongoing communication tools built-in by default. If that was your point, it's well made!

As for bot usage, there is no real way to tell for sure that it wasn't counted, of course, since Nielsen was not behind every 180,000 PC user, 24h/7, and watching them with a camera to see if they were, in fact, using bots or Second Life. The only thing we know is that they measured applications being launched and connecting to online servers. As someone pointed out on Massively, this meant that if several applications shared the same name, they would be registered as just one. Similarly, if someone was quite eager to frustrate the data Nielsen was collecting, they could have created a bot application, name it "SecondLife.exe", and leave it running on their PCs, sniggering with glee as they perverted the data...

Now, Nielsen is not exactly a start-up company that was born yesterday. Any reasonably good statistician knows that extreme cases in the data set ought to be discarded if they are completely out of the range — and are isolated cases pointing to a malfunction, an error in measurement, or, well, plain old cheating. So if out of those 180,000, say, a thousand were launching bots called "SecondLife.exe" and letting it run 24h/7, that would show up on the data set and would be discarded. I have no reason to believe Nielsen would do anything less. Even if the average is one hour and a half per day, extreme cases of 10 or 14 hours being logged in might have been discarded as "fraud". On the other side of the coin, if someone launched a bot just, say, 4-5 hours per day — while in reality they were watching TV — that kind of measurement would probably fail to register on Nielsen's dataset. But that would mean mostly malicious intent in 'doctoring' the data. Now, malicious intent is not a prerogative of Second Life residents; it's uniformly spread among all human beings (at least, statistically speaking). Since Nielsen measured 100 PC games/applications, it's more than reasonable to believe that a similar percentage of cheating would be found among all other games being tested, as people tried to artificially push up the numbers for their favourite games. But that would balance out in the end: sure, all numbers would be slightly above the real data, but would still be able to compare the numbers.

On the other hand, the percentage of cheating has been studied and documented; the authors of Freakonomics, who tie the reasons for cheating to incentives, claim that the number of cheaters is usually around 3-5%. Put into other words: if someone would be paying the 180,000 control group to push the numbers up for their favourite game, there would certainly be more than 3-5% cheaters; lacking strong incentives for cheating, the probability is very high that the number of cheaters is below those 3-5%. So, if you wish, the number quoted for Second Life might be off by 3-5%.

The other numbers are different. The sources (except for Alexa) don't say how those numbers are actually calculated. Often they quote officials from the companies who did some measurements. Alexa is fully automatic. In the latter case, we know that they will obviously count everything that logs in to the respective website — but will probably never see the results of the API calls. The non-Alexa numbers might be officially presented from the companies running those services and I can only imagine that they will include API calls too — after all, a logged in user is a logged in user, no matter what tool they've used.

So you might be right about the Twitter numbers being too low, since they come from Alexa, which has no concept of API calls whatsoever. All it can measure is page hits and do some maths about it. API calls would not be "indexable", unless, of course, the Twitter techs make it that way (if I'd worked for Twitter, I would certainly do it that way!). The simple answer is "we don't know". Twitter also doesn't publish statistics. Nielsen has just published some numbers saying that the "time spent per visit" goes from 5 or so minutes up to 20, depending on age; they don't say how much, on average, a user "visits" Twitter every day.

As for your comment on the time it takes for satisfaction, it's a philosophical one :) Mixing up "efficiency" with "pleasure" is a complex thing, and depending on a personal mindset, the answer shall be different. Short-term pleasure tends to get quickly boring, but while it lasts, I can imagine it might lead to overall higher levels of satisfaction: thus this would mean that a service would be quickly adopted by a lot of people, but they would go away relatively quickly. By contrast, things that get a long time to give "satisfaction", requiring a lot of investment in time, focus, patience, and overall commitment, will give satisfaction for "achievement" ("I finally managed to understand how this works!") which will last a long time. But the reverse side of coin is that they will have an extremely high learning curve, which will mean that a lot of people will give up "too soon" (ie. before reaching satisfaction) and never come back again.

As said, this is a philosophical/psychological issue. In this ADD-ridden world, we see way more examples of the first case (quick satisfaction levels, lots of excitement during the time the satisfaction lasts, an exponential growth in adopters, but the "fad" dies quickly) than on the second one (very slow learning curve, no satisfaction until quite late in the process, a very slow adoption rate curve, a churn rate showing a high number of people who never reached satisfaction — but very long-term adoption with constant satisfaction of those that managed to reach a certain plateaux of know-how required to find the service interesting). But you can apply the same reasoning (which is really about how humans react to stimulation and incentives) to completely different things, like, say, relationships. As more and more people in our society go for the "quicky" model of intense satisfaction for short periods of times, they soon tire of their partners and are unable to form long-term relationships (or consider them utterly impossible, from a purely conceptual point of view — "there is no one that can please me for so many decades!").

But, alas, I'll leave the philosophical/psychological debate for another day :) I will just agree with you that Second Life's model of "high learning curve, slow adoption, but faithful long-term users" is a bit at odds with the current trend in our society, and thus I'm actually surprised that 15.5 million people managed to sign in at all, and that a tenth of that still use it regularly...

The values are average per user. That means, yes, that the average SL user is logged to SL one and a half hours per day, every day. While the average Twitter user just uses it 5 or 6 minutes per day. Note that both statistics only take into account active users!

Also notice that the above statistic also includes combined usage, ie. it means that some people might just have SL turned on for hours on end but be always following people on Twitter AND Facebook AND Friendfeed, or even watching TV and sending text messages on the phone, and so on, during the same period. That's fine, as statistics go! After all, it's not as if we physically "shut down" ourselves while we're logged in to SL; e.g. we're still breathing ;) Multitasking just will show that overall, our minutes spent on all those things will be added and might come up to more than 24 hours a day.

BTW, you can use always Ctrl-H for chat history in SL, you know ;)

There's no question that for pure information retrieval and information sharing, right now, nothing beats the World-Wide Web. Which is hardly surprising: it was designed to make information retrieval and sharing easy. And as a nice side-effect, thanks to Google and other search engines, searching for information is also incredibly easy, even, well, if you get thousands of related links and need to browse through them to find the information you need; but it beats going to the nearest real-world library and search for the same information, of course. Also, as you mention, there is way more information on the Web than on any library on the world! So I find it very hard that virtual worlds might ever beat the 2D WWW at that (I could be wrong though!).

Granted, use a hammer as your only tool, and every problem will look like a nail. This just means that we're so used to the Web to provide us with all sorts of solutions that we "prefer" it (due to our familiarity with it), even when it's not the best tool for a particular service. But that's just mental training. A typical example: you can play role-playing games with a book, jumping from one page to another, rolling dice, scribbling on pieces of paper. But playing it on a computer is far easier. The book is just not a good medium for that. Nevertheless, during a certain moment in time, since people were so familiar with books, they thought that it might work. However, the sequential nature of books does not lead well to a model that relies on pretty much random browsing. We quickly abandoned those kinds of books and used a technology that leads better to "random browsing" (in this particular case, computer-generated 3D games :) ).

This is not to say that virtual worlds will, at some point, replace or surpass what we're doing on the Web right now. Just that some uses of the Web are "clunky" and "unnatural" and will definitely migrate to 3D (a typical example being distance learning, prototyping, teleconferencing, market analysis of consumer products, etc.); new ones will appear (we might not even know what they are now!); but, as said, a lot of things will be served much better by the 2D Web and will not disappear. 2D-based information sharing will very likely continue to be better served by old 2D interfaces.

The issue about bots is always on people's minds — and LL claims that there are, at any moment, 5-15% bots logged in to SL — while they forget that 99% of the Internet traffic is not created by humans. So when you say that "200 unique visitors come to my side every day", it actually means 2 humans and 198 crawlers, spiders, bots, references, links, and so one. It's staggering, but nevertheless true!

This is another reason why Nielsen is insisting on measuring actual people connected to actual computers actually logging in to systems and see what they are doing, instead of relying on "page views" or "unique visitors", which will show all that insane "bot traffic" on the net. The 653 minutes/week is an average based on real people connecting to SL, not bots (with bots, as you said, the number might be slightly higher). It's also not a number from Linden Lab. By contrast, we don't know how the remaining numbers were actually measured. Alexa, of course, will not have the ability to separate the plethora of Twitter bots (even if they're created by humans, they might just have automatic feeds from blogs and other tools pushing messages to it) from real humans. In fact, as Twitter and Facebook become more and more used by companies and institutions promoting their services and products, the usage of Twitter/Facebook bots will increase and increase. Are they already 5-15% as in SL? I have no idea, but I'm guessing that this will soon be the case...

How true, how true! I'm not convinced either that Second Life "as it is now" will be around in five years, but — starting in late 2010 — it will be a complex mesh of several interconnected grids, with several different specialised viewers to connect to each grid...

Hi IYan :) Yes, indeed, I agree that it's the usual business model: idea first, money next, lots of users next, and (probably!) a business plan last. Yuck. No wonder so much tech ventures utterly fail ;) It's all upside down, and this is not my humble opinion, but a rather considerate one :)

Then again, I guess I'm just biased towards what I've learned. Venture capital to test ideas — specially in a market that is completely new, thanks to an awesome idea that didn't exist before — is a great way to innovate. But... business is business. If your model relies on "free access by users", it's doomed in the long term — unless you're willing to create a foundation or similar non-profit to raise funding constantly for a technology that will make the world a better place. Things like the Wikipedia, Apache or Mozilla foundations are good examples, but there are thousands of similar ones.

Usually my argument is: if the model of growing on top of a free user base in expectation that money will come in somehow is a good one, why aren't Microsoft or Apple doing that? Profitability — and a business model based on profitability — never hurt Microsoft or Apple :)

By contrast, a model based on a huge "free user base" is naturally interesting to report on the media. The media loves numbers, specially if they are "hundreds of millions". It shows people like the technology. Well, people love everything that's free, and that's undeniable! But it doesn't mean that you can survive as a company on a free product.
I'm fine in accepting that sometimes you have to cleverly "invent" a business model a posteriori — you make a good point with Google, BTW — but the trouble is, that model has to make sense. Google's model was to get their revenue stream on an unrelated issue. Imagine that they'd started to charge for accessing their search engine — nobody would have used it, people would just continue to use Microsoft's or Yahoo's instead (who had revenue streams from other services allowing them to provide that service for free).

BTW, Google alleged once that they bought YT just to "buy themselves into the copyright wars about video": in the sense that they were going to make a point (namely, that nobody will dare to sue Google for acting like a non-moderating "carrier" of video streaming — precisely the opposite stance of LL, who just love to interfere with residents' content, thus making themselves liable for what their users are doing within SL). If that's true or not, I have no idea. I guess that at the time, two major sharing sites — Flickr and YouTube — were available for grabs, and Yahoo already bought the first, so Google just had two alternatives: compete with Google Video (and Picasa) or simply drop the race for the social sharing tools (Blogger, for instance, was a bad choice; imagine what would have happened if they had bought Auttomatic with WordPress...). Whatever the real reason was, YouTube was not profitable, and very likely will never be — they're still based on Google AdSense to get some revenue and on some people who pay to create longer-than-10-minute videos. I know that at some point Google might have thought they could, for example, just allow high-quality videos for their paid customers, a way to differentiate user accounts and encourage upgrading. Sadly for Google, their competitors — like blip.tv or vimeo — already had high quality videos. So did MySpace. YouTube had no choice but to offer the same, for free. No, I don't believe that YouTube will ever be a profitable Google company. The best that Google can hope for is that they cut more and more costs as their vast server park grows and grows and the running cost per server drops further...

As for Second Life... it does have the right business model, and Linden Lab is profitable. However, like so many things in life that are unfair, that model, as you said, only works for a small niche market, and I'm not sure that even with the announced "SL Lite Viewer" for late 2009 this will change dramatically. SL is too strange, too different. Even Philip clearly states that SL is not a social networking thingy, although it has all characteristics of one. It's something... entirely different. The more I become familiar with SL, the more I believe that it's a technology still too radical for the world-at-large.

We will need to see the Web 2.0 dot-com bubble to collapse first in order to see SL to start really growing. And that will take at least another 5 years, as more and more social networking websites pop up with inexpensive platforms, all over the place, fiercely competing among themselves for a slice of the "market". I put the quotation marks because for me a "market" of free users is not a market at all...

Well, Prok, we can only hope that Linden Lab does better this fall :)

Hehe yes, I've seen several MBAs being announced like that too ;) And people who definitely believed that, as well. Worse: people who believed that their brand-new, MBA-enabled manager would solve all problems...

Ah well. Real life is full of illusions ;)

Whew. Lots to think about :) I should say, right from the start, that your use of my own poor little self as some kind of "personality" to make a point is a bit skewed. As I keep repeating, about only 200 human beings read my blog every day (the rest are 'bots gathering statistics). My blog rank on Technorati keeps falling. The number of regular readers has declined over the years, continuously so. And so on... but alas, it's your essay :)

I have to admit that this is one of the most interesting essays you ever wrote, for lots and lots of reasons. One is clearing out the idea of human beings as "narrative beings", a concept I only had from Terry Pratchett's books, and which I did mostly disregard as "fiction", even if on his partnership with Ian Stewart and Jack S. Cohenon on Science of Discworld II: The Globe, this thought is echoed over and over again. It's interesting to see someone tie this concept — that we humans are, mostly, storytellers — with the notion (or "illusion", as some oriental philosophies would say) of the self as an invented narrative, which, of course, fits quite well in the overall concept that reality is a perception of our senses — in a way, we tell to ourselves the story of the perception of reality (and of others). I definitely favour that argument, too :) In fact, it has strong and powerful consequences, and, ironically, it is a "theory of the universe" (and not only of the mind!) in the sense that "the universe is a collection of stories about what we perceive". Science is a story, too.

Even more intriguing were the quotes from scientists explaining the notion of self as "fragments of stories" that we assemble to, well, become "ourselves" — but that those same fragments can quite easily be assembled to create imaginary characters in fiction. Or, well, on virtual worlds. Eons ago, I wrote something not unlike that: the notion that our self is a dynamic thing that is assembled from several personality traits and that it can get "reshuffled" pretty easily when you're younger, less so after your teens — except, of course, if you're under the influence about some kind of drug, narcotic, stimulant, or, well, through brain damage and/or surgery. In fact, although I didn't realise that at the time, it's exactly because of this ability of the "ever-changing self" that drugs are able to deal with mental disorders like bipolarity or the more common depression — we can artificially "shut down" some areas of the brain, and become "different" in that way.

Granted, if you start reading oriental philosophy or anything the classical Greeks have written 2500 years ago or more, this won't be news. They always said that the notion of "self" was purely delusional — just a story that we tell and share with others.

Starting from this assumption, it naturally follows that if someone can tell your story well enough, they become you. This is, in fact, one of the most worrying aspects of electronic identity theft: creeps being able to impersonate your self as good as you, and, well, use that for illegitimate (or criminal!) purposes. This is a serious crime. One that is hard to prevent. So, if the authorities already worry about identities being stolen, and incorporate that in the body of law that protects our societies, it's obvious that "identities can be copied" (or, well, roleplayed, since that word is quite well loaded). I missed some typical examples on your essay: e.g. things like Sherlock Holmes or even Charlie Chaplin's Charlot that became stereotypes, but whose "stories" will be immediately recognised by anyone — and we can, of course, use many more examples. Are vampires real? No. So why can anybody (in the Western world at least) define what a vampire is with excruciating detail to the point that everybody in the audience will immediately know what they're talking about?

So, I'm obviously not "surprised" by your essay — just surprised, in fact, about the many ties you found between (apparently) different research areas, all of them pretty much saying bits and pieces, but you managed to bring them all together under a consistent idea. Gosh, I just realise that this is exactly what you said that an essay actually is — bits and pieces, floating around, gathering into the same "story". Nothing is new, just recycled — "newness" comes only from the insight of saying which pieces should be assembled together, and which should stay out of it. Uncanny. Very nice work, Extie :)

Lastly, I always find your ideas about "immortality through avatars" amusing. Oh yes, they're not so "obvious" — the transhumanists and extropians are usually more worried about the "mind uploaded to computers" issue. You, on the other hand, minimise the importance of the technology by itself, and point to a far easier route for "immortality": having other people roleplaying your self (and, after all, what better "machine" to upload your mind to — a human being, which are the best known examples of "mind-running" computers that we know about? And hooray, they already exist, work fine, and we have 6.3 billion of them around!).

The issue you always arise is the "why would someone like to roleplay me?", and, of course, this is where we get religious — or perhaps mystical would be a less loaded word. You seem to imply that only "famous" people would likely be roleplayed by others — thus preserving their immortality. In real life, this is, to an extent, true. Sherlock Holmes, for instance, is probably the archetypal detective that has been mostly roleplayed ever, just because, well, Sir Arthur Conan Doyle was famous and his stories became even more famous. A minor detective on an even minor book of an unknown author might never be picked up ever again. Similarly, Plato has roleplayed what Socrates might have said, because, well, Socrates, even if he hasn't ever written a single word (that we know about), was "famous" — more famous than Plato at least. Examples abound.

On the other hand, of course, you make a good argument that "unknown" people (in the sense of "less famous") might be "easier" to roleplay because there is little known about them, and fewer friends to "fool" (in the good sense). That argument is definitely true; the question, of course, arises:

If you have unlimited abilities to create your own self — either in real life or, well, in virtual worlds — why should you be compelled to roleplay someone's self? That's something I still don't get :) I might imagine one scenario: claiming to be "Charlot" in 2009 might be far better for a performer to get an audience, than, well, claiming to be himself. Elvis impersonators are more "famous" than the real persons that impersonate them — just, well, because they impersonate "Elvis". So there is some good argument to say that famous people (whatever "famous" might mean in this context...) will be good candidates for roleplaying. After all, mentally deranged people are keen to say they're reincarnations of Cleopatra or Napoleon, but never of John Doe, anonymous goat keeper of a rural dwelling on the highlands :)

And finally, of course, I might add some things of my own :) If I'm personally not that keen about releasing so many information about my real self — and God knows I give enough hints — does that mean a) I have something to hide; b) I'm aiming for immortality, as you suggest, by forfeiting the link with my real self, so that someone else might pick up the mind-patterns of "Gwyneth Llewelyn" in the future; c) I'm just having fun roleplaying someone; d) none of the above.

Ha! I wish it were an easy answer :) And, of course, the answer is different depending on the year you ask me :) It might make a whole essay one day, but suffice it to say that by disconnecting my virtual self from my real self, I'm just making a simple statement: human beings are worth by what they say and do (you might say: "the story they tell"), not because of who they are, where they're born, how old they are, what they're studied, what cool friends they've got. If there is a simple lesson I've learned is that I, as a person (and that is true of every human being on Earth, even if most will disagree with me :) ), am worth very little. It's just my ego that makes me think otherwise. Everybody else is way more important than me. However, we tend to "tag" people relatively to our social status, wealth, friendship, knowledge, studies, and, well, colour of skin, age, gender, religion, whatnot. I dislike "tags". I'm just another one of the 6.3 billion human beings in existence — nothing else, and nothing more. My virtual projection into Second Life, the cute-ish red-headed avatar that walks around with a smile, a glint in her eyes, and a flower in her head (has nobody ever wondered why?), is just tabula rasa — take me for what stories I spin about myself, not for my, uh, "credentials" or "authority" that comes from immaterial and transitory things that I might have accumulated elsewhere in real life. These are completely irrelevant to what makes me a human being. And by voluntarily discarding all those "real life tags" I allow everybody in SL (and elsewhere) to tag me from scratch based on what they experience.

Granted, this might have been my reasoning, but it has a major flaw: as time passes by in SL, I accumulate new tags :) That runs, of course, against my original intentions (just an hour ago, I logged in to OS Grid, and the first guy I met there just asked: "Hey, are you the same Gwyn that blogs a lot?" *sigh* There goes my theory!). I'm sure that there is a lesson to be learned there, too. The good news, of course, is that as SL grows and grows, I become less and less relevant, and that is a Good Thing.

And of course, there would be an easy way out, e.g. getting different avatars, different names, all the time, so that I could avoid the tagging. Alas, that doesn't work at all. Imagine a tourist visiting a nice, peaceful, fishermen's village at the coast. She won't make an impression if she stays just a few days around and talks to people. She will be quickly forgotten once she leaves. But if she remains in the village for years or decades, she will be accepted by the community, and, even if they remember that she might once have lived elsewhere, she'll be "part of it" now, and will be treated according to the way she presents herself. In a sense, that's my idea of Second Life, strongly influenced by Philip's own idea of "SL as a country". I'm an immigrant here in SL, but after so much time has passed, I feel that I'm accepted now, and can contribute back to the community as well. Starting afresh every day — juggling among alts — defeats that purpose.

Last year's hardware specs definitely include a "vast sea of possible PC configurations" too ;)

Joking besides, the point is that LL's developer team is totally unable to test all possible configurations out there, since, as you so well put it, not even Blizzard can do that... nor Microsoft :)

On the other hand, increasing the number of developers through sharing the open sourced code definitely allows more and more people to be able to test it for their configuration and submit patches that will work for them — and, incidentally, to everybody else who happens to have the same configuration.

As a Mac user I'm part of the lucky class of computer owners where all possible configurations are well known in advance, of course, so your suggestion would benefit Mac users immediately. But, alas, SL ought to run well on non-Mac hardware too... :-)

Ah... I wonder if you're using this script on a WordPress blog? :)

Because if you aren't, you'll need to download <tt>class-IXR.php</tt>, install it somewhere, and rewrite the whole code to point the paths properly to it. Or, well, use any other PHP XML-RPC packages (there are a lot of them; Incutio's IXR just happens to be small, efficient, and very simple to us; these days, most installations of PHP also include XML-RPC classes, which are probably faster since they're C libraries :) ).

Oh yes, I most definitely will use http-in as soon as it's available (hopefully still in 2009!). LL will slowly phase out XML-RPC — and not abruptly — since their own XstreetSL.com web shop uses XML-RPC profusely, and a hundred thousand people use that every day. That's a lot of code to replace!

Oops!!

Well, nobody reads my blog anyway, I'm sure that LL's more concerned in pushing servers around than in strict compliance with NDAs... hopefully.

@Prokofy, I personally never make the claim that "open source" means "absolutely free", not even that "it always saves costs". It all depends on the specific situation and a specific type of software. Most frequently, companies shrug off the costs of maintenance and support when using open source software, i.e. "who do I call when I have a problem?". If they are able to afford to pay someone in-house to keep running the software for them, then sure, go ahead, use it. If not, outsourcing closed software might be a cheaper solution.

Many of the most popular open software platforms developed by professional companies rely on the business model that most companies out there don't have the know-how — or cannot afford it — to run their open source software, and sell basically "licenses" for doing the installation, maintenance, and customer support. Typical examples are Alfresco (data management, workflow), Zimbra (a MS Exchange clone), Red Hat (an operating system based on a Linux kernel), MySQL (database management), Mono (a clone of .NET), WebKit (an HTML rendering engine used by Apple and Google), or even Boonex's Dolphin (a social site building tool) — but there are tens of thousands of others. None of these companies went broke because they released their software open source. In fact, they're quite well-off, they have millions of happy customers, many of which are quite happy to pay for customer support and maintenance.

The difference in all those applications is that there are always a large group of people that download the software for free, attempt to install it, validate the code, submit patches and new features, organise a forum or similar group-discussion tool, exchange tips, help new users, and spread the happy news (free promotion!). This means that an obscure, small company can quickly build on an established base of knowledgeable users, which will become their evangelists, and push their product to colleagues, partners, or even clients — for free. It's that crowdsourcing work (not only in programming, but also marketing and customer support) that is so appealing for companies who are willing to release their source code.

All those companies with their products also keep a degree of control over their software. Pardon the pun, but it's not a "free for all": these companies rely on selling maintenance and customer support agreements for their software and are naturally more than interested to point the software in the direction that their paying customers want. Getting free testing, bug reporting, patches, and contributions is just an added bonus — often a very important one, since mature code mostly means "heavilly reviewed code", as in: "thousands of people have seen the code and get it to work properly". Small software houses are not always able to get enough people to review their code. Even large ones (like, well, Sun, Apple, Novell, IBM...) benefit from an extended group of code reviewers, allowing them to focus their developers on new and different projects.

So your ideological view that "open source is evil" is totally misguided. You're thinking mostly of the small, one-person development efforts which are here one day, and disappear the next. There are actually very, very few projects that survive that way. Programmers are notorious for their short attention span, and too individualistic to truly share their expertise (specially for free). What they aim is to gather enough "fame" for their project and attract similarly-minded programmers, and — hopefully — keep it going, until, say, an enough number of them are able to create a company (or a foundation) to act as keepers of the code. It takes someone to do that work — and work it is, so it better be paid!

Oh, sure, of course there are exceptions to the above examples. There are always exceptions — edge cases — which will be brought up whenever someone wants to make a point that some open source software survives even without "companies" behind them. These are notorious exceptions in any case. Long-lasting projects, however, are different :)

@kanomi, just remember three things: first, Linden Lab will not "abandon" their main series of viewers. These will continue to remain available for download. Secondly, a whole generation of contributed patches that actually make the viewer work better and faster will be immediately applied to the "bleeding edge" viewer, meaning that at last we won't have to wait for LL to apply a 6 or 12-month.old patch to their code to have it work better and faster. And thirdly, on most open source projects that have a company or organisation behind it, the resulting software is usually quite thoroughly reviewed — just because there are way more people doing so. Many forget that before there was a quality assurance team at Linden Lab, LL relied on volunteers to do the testing for them — there are still Wiki pages showing what those tests are (or were). These are quite thorough. When the independent developers start to notice that their testing and patch submission actually goes through, they might feel encouraged to test even more and submit more patches... and thus making the overall code be of even better quality. That's what happened to almost all open source projects, and it's particularly interesting to see how things like "clones" of existing software (e.g. Mono's replica of .NET) actually become way more robust than the original software, in much less time — just because they get so many more people to tinker with the code, test it thoroughly, and make it better.

I should add, for the sake of completeness, that if you're serious about doing some real work on OpenSim, you should not use the kind of specifications I use :) I'd definitely recommend emulating Linden Lab's own Class 5 servers at least: quad-core CPUs with at least 2 GB of RAM can get inexpensively leased for a few hundreds of dollars per month, and you should definitely get your central servers and the database on separate servers — inside the same provider to keep bandwidth bottlenecks down to an absolute minimum. I can imagine that if I could afford it, I'd have at least 6 servers, all interconnected on a local Gigabit Ethernet switch: a MySQL cluster of two machines; one for the UGAIM servers; two quad-core latest-generation servers with a minimum of 2 GB (but I'd go for 8-16 if I had more users) to run, say, 20-30 sims; and a separate server for tests, maintenance, a support website (where you could do things like remote administration — which is supported by OpenSim, btw), the economy server, an IRC server for people who can't log in at the moment, and gathering metrics and statistics. That will set you back a thousand dollars or so per month — plus, of course, the cost of your own work to keep the minigrid up.

Alas, Tracy, it's not up to me to say what is "junk" and what is "information", since someone's information is junk to everybody else, and vice-versa...

My role is definitely not be part of the "content police" and say what should be allowed or not ;)

Kevin, which of the two PHP scripts gives that error to you?

Wow interesting, spy! :) I'm flattered, since so few people actually bought that device....

@CronoCloud, you definitely have a point there, of course :)

@Blue, what actually surprises me is that I'm still on the top 5% ;) In late 2004, I had 4 times the audience, and ranked third after New World Notes and Prokofy Neva. There was nothing to read about SL :) Now that there are at least some 2,000 blogs on SL — and I guess these are the ones in English — I'm always fascinated how people still have time to read the things I write... I wouldn't have any patience, lol

*goes back to read her favourite fashionista blog*