How could SL go open source and be profitable?

While an open source client might be great in terms of functionality and reliability, an open source client would kill copy protection in SL. I could simply modify my client to save every texture, sound, animation, and prim configuration that I was interested in.

- Ace

Once Linden Labs fall out of love with what they're doing they'll more than likely go open source but until that time I very much doubt it.

Equate it to having a child under teen age that you nuture and love,once the child reaches teen age then it's time to let go and let outside influences creep in.

Richard Garriot was the same with Ultima Online when he sold it on to EA, CRS Software are very much like LL in the way they fiercly defend the developement of their world.It's an indication that Linden Labs are not JUST in it for the money.

Of course we like to think of Second Life as hugely varied but this is'nt strictly true because its really influenced and guided by a relatively small number of people.A move to open source would be good AND bad for many many reasons the main one being that once that move is made then your going to loose this sense of 'family' that we have now.

It's good being here at the stage we are at now,try playing SWG by Sony or Lineage2 or WOW where the devs dont really give a toss what you think and you'll see the difference.Second Life isnt just the way it is because of the amount of freedom it allows,its this way because of the things Linden labs allow.

You know what? I think they might actually care.

You can already do that now, if you're so inclined, it's just a bit messy to grab and disentangle the client memory, or to intercept the calls to the graphics card, the sound streams, etc etc.

Digital Restrictions Management has only ever worked against the man in the street, never against someone who understands how computers work.

There is only one surefire method of "protecting" income from those resources, and that is to lower prices until overcoming the protection is too much trouble to bother. That applies in SL, just as much as it applies to music and the RIAA dinosaur.

Most people log in to SL for a bit of fun, and if prices are completely reasonable compared to the weekly stipend then they won't even consider looking for a means of overcoming restrictions.

But for those who think that they have a right to make a killing through restrictions and inflated prices then yes, the writing is on the wall. And good riddance.

Yeah.

One of the reason's Jarods comments about the protocols seems strange to me is that I'm just waiting for someone to reverse engineer the protocols.

It is possible. The client has to decrypt in order to utilize the information, so it's in memory just waiting for you.

This is one of the reasons why SL is going to open source the protocols at some point. Because they *have to*.

Indeed. I imagine that nobody's bothered mainly because SL is such an easy-going place, with little competition and nastiness going on.

That contrasts sharply with some other online places, EverQuest in particular as an extreme example, where the numerous hardships of life and the economy of scarcity resulted in a lot of hacks and external data capture systems to try to overcome some of the pain.

It's pretty much the same in all walks of life, even outside gaming and virtual worlds. Give people a strong reason to object to the status quo and they'll find a way to make life more pleasant for themselves and their friends. Give them a pleasant environment to start with and enough empowerment to let them feel that they're more than just cattle, and they'll stay within the lightly-fenced bounds of the playing field.

I would argue the problem here is the means, not the motive. Regardless, another conversation for another time.

I think the fact SL will go open source is inevitable, and the higher ups seem to agree with that sentiment. The real question we should be asking isn't "will;" we should be asking "when."

And, in that light, I would say when the userbase is large enough to be self-sufficient. That's a while off, yet.

Oh, and good points here.

I think that many people in here are ignoring one of the fundamental shifts that will take place if and when SL becomes a metaverse.

"Open-Source? What about my uber-widget ver 5.2, I spent so much time developing it and now people will have ways (maybe) of hacking it and depriving me of money that's rightfully mine. "


1. Once Secondlife becomes a metaverse, people will be much more concerned with buying actual widgets, not virtual ones. They'll want to use the interface to shop at Amazon, at BestBuy online, order a pizza for delivery or a porn movie that they can watch on a screen in a 3-d microverse while re-enacting the scene with a virtuality of the stars of the film. Virtual widgets will still be an interest, but a side one, used to entertain, to pass the time, to draw customers in, or created for the sake of creation itself.
Once the real-life retailers arrive, their business activities will dwarf the virtual-goods markets. The largest demand for virtual goods will probably be in clothing/appearance, and in virtual goods for metaverse games; that is, games like RPG's that are run on their own servers.

2. Along with the surge of real-life commerce conducted virtually, people will be more interested in earning or spending real dollars rather than virtual dollars which can maybe be exhanged for real dollars (value dependant on fluctuations in both currencies and strongly tied to confidence in a software product). The Linden virtual economy is just a testbed for this; if and when "metaverse" arrives, the "Linden dollar" will either become an ingrained means for cash exchange (like Paypal) or it will fade away in favor of a different system.

3. Once new grids pop up, especially ones not being run by LL, we'll need a new permissions behavior for virtual goods. The behavior will likely be something like "Tied to grid, (y/n)", "Allow owner to tie to (x) grids" The idea is that a creator can determine whether the use of their objects can be tied to a specific grid, how many grids it can be tied to, which ones, and also set it so that the new owner might, say, be able to tie the object to any 3 grids of their choice. Two way authentication would have the benefit of both making the items harder to hack, and providing for the security of game-system grids, whose owners take steps to NOT allow their grid to recognize hacked objects. Even if you hack and copy an object, that grid won't allow you to bring it into play. You'll have to resort to using it on whatever pirate grids exist, assuming that they provide any functionality for your hacked widget.



I would be interested in seeing a plan where:

Client and server software sell for a nominal price, for nonprofit home use, client used to connect, server only for offline (LAN) use. "Enterprise" edition subscriptions for the server used to take advantage of meta-space; subscription rates varying depending on whether self-hosting or paying LL for hostspace.

Paid subscriptions not only get the ability to host, they gain admittence to support forums that non-paid subscriptions do not get. Anyone may view the forums, but to post and ask for help, or use the search feature, one must be a subscribing member.

LL aims to achieve profitability by providing tech support, upgrades and hosting in return for subscriptions, and also by creating a metaverse money exhange (Like Paypal) for secure trusted transactions at retail websites. Their financial services are not mandatory, but may receive a certain level of trust because the transactions are being handled by the same company that designed the interface.

I tend to agree with your points 1, 2 and 3, Unhygienix, with one proviso: "everything is transient".

Yes, these things will happen, but once a revenue stream is found, it's not a revenue stream for life --- a lesson that the RIAA (and authors and artists) still haven't learned.

We're on a technical voyage into the unknown, and if there's one thing that's certain it's that things will change, all the time. And that's one reason why I have no time whatsoever for people who want to protect any particular virtual resource as if it had some inherent value. It doesn't. It merely has a value in one point in time, and times change.

That was the "Yes" part just above, Unhygienix, this is the "No" part ...

If LL were to adopt anything like the commercial sublicensing/concessions system that you described, it would no longer be in the game of spearheading the development of the future open metaverse, because it would have commercial revenue streams to protect. And it would do so, it's in the nature of business.

The metaverse initiatives would then have to go it alone, in the regular open source community.

This sounds to me like you're setting the stage for a second Dot.Com bust centered around Second Life going open source. I find that to be a very interesting take on the matter - though I don't want to put words in your mouth.

What do you think? Will the upsurge in demand bring that kind of outcome?

How about reaching first for a 2nd dotcom bubble, before looking for a 2nd dotcom bust.

Fine. Guess we can't speculate *that* far into the future just yet.

Btw, I have no doubts whatsoever that there will be a 2nd dotcom bubble/explosion, once an open metaverse is born akin to the open Internet of a decade ago.

And that's why LL must stay in the lead now if it wants to be huge, disregarding everything else like point-in-time revenues and silly protective measures.

This will be a 500 billion dollar industry in a future that's not too far away. Those who base their planning on current business and product are destined to be late to the party.

I thought about this sometime last year...

it's quite probable and likely that LL will head in the OSS direction. "Free" is something I don't think they'd like to get into too much.

On some technical points -- why an asset server?

I understand that in the here and now it's sort-of necessary;

in the future however, wouldn't it be more effcient to store a users' inventory locally and merely upload themselves or request objects from inventory to rez at the server they are connected to?

It would only require a minor upgrade in upload speeds to transmit and disseminate the 3D vector information which is already rendered client-side.

Then LL could act as the initial host company and obviously control the sever-software license. Perhaps even sell a certification program to developers a la MS/Redhat/etc

I'm a fan of local inventory and a low-scale local server to allow the client to browse their inventory offline. Also, with local inventories, the server software could be upgraded with malicious software detection plugins/scanners etc.

I dunno... daydreaming.

As an added note on the inevitability of SL becoming OSS:

Consider that the goal of business is efficiency, not profits. You make profits by being efficient.

The problem with big business is that they become so large that efficiency is something they begin to lack, so they start spending their profits to continue making profits... and that's where society is stuck right now.

However, no matter how big the company is -- it's goal is still efficiency, even if obscured by profit motivations. There are many examples of some big businesses overcoming size issues through efficient franchising... which they will continue to do.

But with increasing efficiency, we're moving to minimal cost production. Profits are made by making cost of production lower -- hence evil practices such as sweatshops and nobel ones such as FOSS. In the software industry, minimal cost production IS FOSS. This means that software in every reach of the world will one day be free -- of course we'll never get rid of the human component so we may find another country to exploit cheap labour from ---

but the point is that LL will reach a point when it's no longer feasible to hold onto all the costs associated with keeping its technology proprietary. It will become more efficient to guide the development rather than be the development.

When that will happen, who knows but it may be longer than we think.

Simple answer: Yes.

Architecturally, this is nothing new: think of DNS. You are authoritative over the data that you define, yet unless you are a mega corp you cannot possibly cope with everyone in the world asking you for your current state on every page access. That's where an independent, distributed, hierarchical system of caching servers comes in, allowing you to retain authority on the flimsiest of machines while a mighty backbone keeps you from instant death from the popularity of your domain name.

An open metaverse will work the same way, simply because there is no real alternative.

Everyone will in due course have their own "asset server" ... in the sense of some sort of local database for their personal resources. The key to success isn't this authoritative asset server though, but the caching asset servers which other people's machines will query instead of your own.

One can expect the major resource providers of an open metaverse to be running huge asset caches to make all this hold together and keep you from meltdown, and it's those caching servers that will be the main reason why you'll be paying them a fee.

No doubt that's where LL will want to be in due course.

Just something to point out:

Open source does not mean 'free of charge.'

The GPL explicitly allows you to sell GPL'd binaries, the only condition of the GPL is that you attach the source code used to build it, with them. Of course redistrobution is a problem; but open source does not have to mean free of charge.

In terms of the actual topic at hand - LL would most probably survive as a service orientated company, providing art teams, server hosting, and general consulting - as well as probably sit in a position as the architecture review board is for OpenGL. (As in, general controller of where the protocol / architecture is heading).

This doesnt stop them from hosting their own SL-like grid on the larger grid either.

-Adam

Oh, I would expect their involvement to go *far* beyond that!

You see, in a free and open metaverse, where anyone can attach their own world in the same way as anyone can hook their own webserver into today's Internet, mere mortals like you and I wouldn't have the megabuck machinery nor bandwidth to support 10,000 people coming to watch the football match that we're hosting --- it would be instant line and PC meltdown!

That's where the corporate providers come in, with their huge server farms running world caches for the metaverse, so that we can define worlds and events freely on our local puny (but authoritative) machinery while they bear the brunt of the brutal assault from all those converging spectators, and earn $$ for it.

LL is well placed to fill such a role, although I am sure that there will be as many competitors lined up to receive our cash as there are ISPs or website purveyors today. We'll want to pay them because it's the only way our worlds could survive going online ... unless something miraculous happens and home PCs suddenly get a X 10,000 speed boost.

In addition to that fundamental role in underpinning the metaverse, they'll probably also continue to provide their own world(s) as now, but that's actually quite a separate issue.

Just for clarity, I'd also point out that GPL is not the only open source license. Also, I'm not sure if the GPL has ever been tested in a lawsuit. Does anyone know if it has?

Anyway, Open Source != GPL

The two dimensions that are often discussed with open source are price and freedom.

That is, what does the source cost to get. And what can you do with it once you get it.

This is general knowledge to people in this thread, but I make the point for people new to open source that might be reading.

Many really successful open source projects are not GPL. Eclipse and Struts come to mind for me since I'm a Java head.

In a really complex system like SL, it might make sense to license different pieces of it under different licenses.

The GPL being tested really isnt worth worrying about: The GPL is a grant of rights, not a restriction. If the GPL was invalidated, then your ability to use GPL'd code would be revoked, not given free will.

That being said, it has been tested a few times (particularly a lot in international courts, eg germany) and in the US itself, no-one has tried challenging it (since it is fairly well worded, and disputes tend to be very one sided.)

As far as 'the open source license goes' - no it's not the only one, but it does tend to be the largest and most commonly used by a wide margin. At last count, freshmeat had 95% of projects GPL, 3% BSD (2 and 3 clause), and only 2% in other licenses.

Again, with that said, I've released a bit personally - but always under a BSD license. I personally dont like the viral nature of the GPL, nor agree with Stallman's ideals (especially not after reading his GNU manifesto).

-Adam

The GPL is probably pretty damn strong. In my days at MSFT they were RIGID about employees not working with any kind of open source code, precisely due to the threat that if anybody could coherently tie anything having to do with OSS to anything having to do with MS products, there was potential for serious trouble.

Some detail: all dev types, particularly people focused on .Net, were brought into meetings with MSFT legal counsel (as groups not individually). The head of MSFT legal himself (interestingly enough a man named Nukem) did a round of presentations instructing us not work with the source of Linux, Quicktime (which had previously gone open source) and so on. Java was not on the bill since that is not open source (something many people are confused about). Essentially, in no way shape or form could an MSFT technologist demonstrate any value whatsoever of OSS. It was garbage and we didn't want it in our products, so to speak.

Naturally a great many of us, as professional technologists, had a problem with this. We got around it in our own ways, many by leaving.

Well, I still think LF's direction is what LL has in mind.

Given havok, speedtree, open sourcing the servers may be very problematic. However, open sourcing the clients seems likely for many reasons, among them are:

a) people are probably going to reverse engineer the protocols anyways
b) more people can help with bug fixing (for example, driver issues)
c) more people can help with new features

However, I think they are under-estimating the probability of a back end open source solution based on something like Crystal Space (forget croquet). Hopefully they have some kind of backup recourse which depends on one of or all of:

a) micropayment transaction system with paypal type fees (except smaller)
b) Google type searching credibility
c) Hosting (basically what they do now)
b) payment rights to hook up to be part of the contigous 'main grid'

I honestly am not sure they can compete on hosting, and do people really care if they are hooked up to the main grid?

They have a good chance on b) but the search engines might get involved if it becomes popular.

I think their best chance lies with a). No one does micropayments like LL does (hell! no one does micropayments, really) and as controllers of what gets into the tree, they can ensure that their place as micropayment king will remain.

Therefore, really, I think in the case of something going awry and CrystalSpace develops dynamic streaming capability, their best chance is to solidify their payment transaction system.

I should repost my SL should go P2P post which I first did when I joined SL. I think in a lot of ways the ideas remain very relevant.

This may need some interpretation to be comprehended:

"any value whatsoever" == "any financial value to Microsoft whatsoever"

As far as simple technical value is concerned, and even more importantly, as far as compounded exponential monotonic incremental growth is concerned, they don't doubt the value, or they wouldn't be so scared shitless. If there's one adversary I really would not want to go up against, it's Mathematics.

Not disagreeing with you Adam.

I only brought up the issue of GPL because it has a reputation at least for being extremely anti-business. Just because a lot of things have been released under GPL does not mean it's a good thing. And it doesn't mean that those projects are achieving the most that they could.

GPL has been around a long time, but open source did not begin to flourish until recently. I believe that the growth of it is in part because of alternative licenses.