the eBay-ization of Second Life?

I constantly see a lot of eBay lessons used in Linden Labs plans, which is not surprising given that eBay is an also an enabling platform (eBay doesn't buy and sell stuff, the participants do) and given that eBay's founder and VC are investors in LL.

eBay does not charge consumers to use the site. They only charge sellers a small fee to list, and a success fee upon successful sale.

My question to all you bright minds is this: do you see LL getting to a point where they no longer charge for tier, but instead make their money from those who sell through the platform?

The model has worked well for eBay because it means there are many more buyers, and thus sellers have a greater chance to make more money. Sellers don't mind being responsible for eBay's revenues.

Likewise, as a content creator (or service provider) in SL, while at first this new payment structure might look like an unfair and painful tax, it might have the potential of increasing the popularity of SL and thus increasing the revenue potential for all of us. You would be subsidizing your customers, but potentially make more money in the end.

Personally I think there would need to be limits... maybe free usage/land up to a certain point and then paid levels. But I am interested in hearing your pros and cons. I also want to think through how it would affect the land trading market.

[PS -- why might this be necessary? well, I am a little exhausted by residents who complain about money/stipend because they don't realize that goodies don't come with this "game" but that you are buying goods/services from other residents. Will this consumer behavior/attitude problem remain a problem until it becomes much more obvious that SL is an eBay-like platform?]

Considering how hugely succesfull this model is, it makes a lot of sense to apply the idea to SL (and hopefully other competitors connected with it) - not only because eBay's founder is invested in LL. Its is a model which is very attractive, mostly because consumer get the feeling, that "it costs nothing" - which is wrong of course, but still a nice feeling.

Currently I see SL as still "way to go" for such a model. The intensity of business and/or system performance has to be raised significantly to make this a viable model. Just look at the numbers:

A server now costs around 200 US$ per month. Given that this a kind of managed application hosting of some complex piece of SW, there is some administrative overhead and in the long run some license costs would be needed for covering further develeopments etc. this is not too bad.

There is a ratio of 30-40 residents/server if you look at the statistical analysis Philip is so fond of. No, this won't change with Havok 2! Its not a performance limit, but a result of the average relations between land/server and land/user.

So LL needs to make earnings of around 5 US$/month per user (more than 1,200 L$).

If this would be a commission for transactions, and we would assume a 5% commission (too much, but its makes calculations easier), we are talking about spendings of 80 US$/month (more than 20,000 L$/month) per user.

At 2% commission it would be spendings of 250 US$/month (more than 65,000 L$/month) per user.

Numbers would look more promising of course, if there was more land/server, which should be possible with better servers and software. And yes, there is Moore's law. So maybe in some 3 - 5 years this model would look much more realistic.




But please lets not forget, that the web does not only consist of eBay. And I see SL (and similar systems) more in line with "The Web", not only with one applications category. So, there might be other successful revenue models, too. If you look at the blogosphere for example, currently the fastest growing source of new content on the web, and another way for people to express their (verbal) creativity. There are a lot of "free" services here but also some very good subscription based ASP solutions for those who want "a little more".

"Free" solutions are usually paid for with advertising. And yes, there is still some content that is actually "paid" for - even on the web. I guess all of these models will coexist in the Metaverse, too

I have heard that Sony Online is moving towards this model.

They are going to offer a free MMOG and then they're going to simply handle all the person to person transactions.

hmmm... good for you to actually put some real numbers at work pham. I should have done that as well. So you are saying that LL is launching a new SIM server for every 40 users that joins SL? That would indeed make it expensive to support at current rates.

Other problems I see:
For example -- because servers are still relatively expensive to manage, if there are no tier costs, how do you divide up land? What prevents a user from trying to hold 2 or 20 sims? One control mechanism would be to have free tier up to a point... maybe 1024m2 or 2056m2... and then charge.) But your land/user/expense ratio still makes this difficult, I agree.

I also think a scheme like this would be quite difficult with a local currency. Does the eBay model only work because everyone trades in dollars? So sellers can be paid directly through paypal?

. . .
re: advertising, I guess that would be a viable alternative if LL came up with an ad-server system and offered you free land or possibly a small revenue share in exchange for putting up ads (which they sell) and creating dwell/clickthroughs

Some other numbers to consider, just for arguments sake:

The new LL Opteron servers run 2 sims, 80 people max, instead of a single simulator.

LL has over 1000 servers - way more, so they probably get a reduced price over what I'm paying per server.

Cool analysis, though, Pham!

-Flip

Hi Flip, first let me say, I am sorry for answering this late. I could not log in for a few days.

And ... Yes, obviously LL will get a better price for hosting 1,000 servers, than you (or me) would get for hosting 1 or 10. And obviously high end servers will be able to handle more than one sim this year with the same performance these sims ran on the hardware current two years ago.

Still, I think the price/performance ratio has to be around 10 times better than it is now, to make the ebay-ization feasible. Which will not take very long as we know the IT business - but wont happen next year either.

Sorry for my late reply also, only saw it due to the new reply. Most servers currently run 2 sims. The new servers run 4.

Just have a look at the statistics that Phil if obviously so fond of. Most of us seem to take them for "just a bunch of numbers". And admitted, you dont get much help in interpreting them in those bare excel files. But analysing them for one or two hours actually gives so some interesting information about the "inner workings" of the SL economy and the major trends.

I am not sure if it would need the Lindens to set up such a system. Most of it could be done fine with some LSL and an external adserver. On the other hand, if I look at the GOM example I surely would not want to invest a lot of development work in such a solution

What are the specs for those servers?

I am not sure Blaze, am not very savvy with that kind of stuff. I was told that the despite running 4 sims, it is expected that the sims will have a 10-30% increase in performance. Maybe someone else will be able to help out more with the details

The problem I see with trying to run Second Life like Ebay is that it is more multi-purpose than Ebay. Second Life has a huge number of purposes, with selling only being involved in some of them. There are many applications for Second Life that do not involve selling at all, but they still would possibly be in need of land and prims. One such application would be using Second Life for long distance meeting and presentation. Another would be people wanting to work together to make an entertaining environment to relax in without selling things. If you focused only on some % of sales, you would be putting an unfair amount of fees on those who sell thing to support the maintenance cost. I think in this environment the fees should try to distribute the maintenance cost across as many applications of SL as possible. Land seems like a logical simple way of doing this. A sale % fee might work in great moderation, but I think it would fail miserably as a stand alone model for fee collection.

I aggree! But I guess any model is kind of flawed as long as we have a single supplier that offers basically one model (with variations).

What Forseti was hinting at (if I understand him correctly) was the attractiveness of a model, where players get more resources than they get now (seemingly) for free. That is, what made eBay and some other platforms big. And it would definitely help SL (or another, similar platform) grow faster, because it is, what a lot of users are expecting on the net of today. And something like that is a viable alternative for SL in a few years.

But IMHO there definitely will be other alternative models for refinancing the cost of hosting such a platform. As I said: the web does not consist of eBay alone and other platforms are offering other refinancing models. The same will happen in the Metaverse.

oh i agree pham. I was just focusing in on LL pricing and revenue generation structures.

dark, good points

pham, i need to think through the notion of multiple pricing schemes -- because in this case it is coming from one company. Are you talking about different pricing for different types of customers? I suppose that could work, although things like that can often be gamed, and the complexity for LL would be greater. Interesting... will ponder. Hopefully didn't read your posts so fast I misunderstood. I'm hitting the sack now, lol.

thanks for responses all

I was just discussing it in principle - along the lines of the analogy Metaverse/Web. And I am not sure if it would be easy - or economically sound - to mix different pricing structures with only one supplier.

I was thinking about a (far away) future where we might have interconnected grids - maybe to far away

On the other hand SL already has different pricing models. The lifetime membership already is an attempt to bring in customers who are not willing to directly pay LL for costs connected with their account with a monthly subscription or via tier payments. I dont know what role this plays exactly in LLs business model. But I guess LL is expecting these customers to either generate revenue in other ways or bring in more paying customers, for example by raising the earnings of content creators with bought or earned L$...

Hold on, Pham. What lifetime membership? Have I missed the announcement of some big new opportunity ? I thought entry to that was phased out in the dim mists of prehistory ?

Oops. I don't want to raise any false hopes. What I meant with that - and used the wrong words obviously - was the "Basic Account" which once did cost 9.95 US$ and is free now

My only excuse is, that you indeed get a lifetime membership with that - with very limited amenities.

Curses. Hope flickered there for a moment