I was reading August 2007 Issue of Wired Magazine tonight (the August issue, 15.08, it's not online yet) article on Second Life called Lonely Planet - Second Life: It's so Popular, No one goes there anymore. How Madison Avenue is wasting millions of dollars creating ads for an empty digital world - article by Frank Rose.
Rather than focusing on the what some consider the failures - I'm saw one paragraph that seems to explain what's really happening with Second Life which starts on the bottom right of page 143 (I will have to hand copy this as it does not appear to be online yet):
"..The simple model they all grew up with - the 30 second spot, delivered through the mass reach of television - is no longer working. And there are two types of people out there (experimenting with Second Life) a small group that's experimenting thoughtful, and a large group that's trying the next thing to come through the door." Second Life appeals to the latter - the ones who are afraid of missing out, who don't consider half a million dollars to be a lot of money, and who haven't figured out (or don't want to admit) that Second Life is less than the bold new frontier it appears to be.
"For people who've grown up in analog, Second Life is not that hard to understand," said Rishad Tobaccowala, CEO of Denuo, a consulting arm of the global ad giant Publicis Groupe. "I have a store in the real world; I have a store in the virtual world." In contrast, the kind of digital marketing that actually works (in the virtual world) requires a conceptual leap. Successful online marketing is targeted and specific, like direct mail - but it's direct mail in a fun house, where the recipients can easily seize control of what the mail says, where it goes next, and how it gets there. You need to know how to buy up keywords to maximize search returns, how to make the most of recommendation engines, how to use the viral potential of Web Video, how to monitor what is being said in blogs and message boards, how not to blow it by trying to be deceptive. Building a corporate pavilion in Second Life doesn't require any of these things. It's simple and it's obvious."
I added the words in blue italics in the paragraphs above, BTW. While successful online marketing statement wasn't really directed to Second Life - I think the basic point being made is the failure of Businesses in Second Life the article cites (that you'll be able to read online or on the newsstands soon enough) such as Coke's Virtual Thirst Pavilion and the NBA has more to do with not engaging the audiences in Second Life than anything else.
To be fair and honest, the Second Life platform is immature in that it can't handle large numbers of visitors on any spot, the grid is down way too often and the interface is still clunky …. but, generally speaking, what's getting businesses to fail in Second Life is businesses copying their physical store in a Virtual World, rather than really coming up with something creative (even though more of the brands who have entered Second Life think they are being creative, they aren't).
There's also a lack of good metrics but that is being addressed via new analytics platforms like Code4Software.com V-Tracker - a full blown analytics solution that takes only a couple of hours to implement on any island.
So, my take is - your not wasting your money by investing in Second Life, provided your willing to think outside the box, look at branding more than profit for now, and think in terms of learning that will pay dividends in the next two or three years.
It could be argued that brands that are closing up shop simply did not have an approach that would work well for the audiences in Second Life - and the article points out that people who build islands are conceptually lazy, they just wanted to try the next thing, without really understanding what they were trying to do in Second Life and why.
But for every 7 or 8 businesses that leave because they did not achieve what they wanted today - there's' two or three that stay and end up being smarter and stronger than before - I would look at those companies/brands/islands and try to understand what is working vs what is not and then, for those businesses that left, I'd suggest coming back and trying again, in Second Life, with a better concept and better metrics (which are now available via V-Tracker, etc).
just another view 
