Second LIfe hosting Croquet? (long ramble)

David P. Reed wrote:
+1

Cheers,
   - Andreas

I think that the original thread was regarding the difficulties involved in bringing content over from SecondLife into Croquet.  One of SecondLife's strongest aspects is the large amount of content that has been created for it, one of Croquet's biggest challenges once things mature is to begin building a content base around it.

Porting assets from SecondLife into some other system at this point is not technically difficult, the only real thing protecting client-side assets is a promise from the server storing the material and the client displaying the material.  This is in opposition to the ownership system that has been setup to promote creation of such material.

The difference between the promised and technical difficulty in examining and porting objects has been an incredibly sore spot and caused a number of flamewars much hotter than the one I am reading here.  Often times this turns into a philosophical argument between the "hackers" (original meaning) and the entrepreneurs.  The CopyBot utility was a modified client that had the ability to make permanent copies of object appearances and temporary copies of avatar appearances.  LindenLabs originally tried to minimize the damage of the application (many of them are hackers at heart), but eventually banned the application and others like it.

Which brings us back to Croquet.  SecondLife has a vast amount of flora and fauna available, which Croquet will soon be feeling a strong need to have.  If we are to import assets from other services, we need to be extremely careful in how we end up doing that and where we get permission for doing so.  Otherwise we could find ourselves in a much larger maelstrom than we realize, like the stunned CopyBot authors when the boycotts began.

Sounds analogous to the windows vs. linux debate.  Some think linux should make itself more consumer-friendly, but I think most linux developers would rather focus on software quality than on market share.  It might be unproductive in early development stages for a free, distributed system to spend much time on interoperability with non-free, server-centric systems.  Better to focus on quality, reliability, and simplicity.

-David Gleason

Creative works fixed in some tangible form are subject to copyright
laws.   Anyone who copies them without permission would certainly be
liable for infringement, following the rules there.  We could have some
wonderful debates here about "tangible" but I think that is not really
relevant to Croquet, and I'm generally comfortable with the idea that
fixing a model in computer storage is fixing it in tangible form.

That said, the creator tends to own the copyright (the "Right" which is
then licensed).  Not being close to Linden's contractual terms for using
the site, I nonetheless doubt that Linden "owns" the right, nor is it
granted an exclusive license, I suspect.   I suspect the creator retains
rights to the content when it wanders outside SL.

That would seem to make the issue simple: if you want to use a
copyrighted object, be it an avatar or otherwise, get the rights to it
from its creator.  

Now there are lots of cases where licenses are mplied.   For example, by
granting a license to an algorithm under GPL, it is understood by
precedent and necessity that you are granting a limited license to any
patents that you have the right to license to the algorithms embodied in
the copyrighted code, to be used by anyone who uses the copyrighted code
legally under the GPL's terms as part of that distribution.  (no broader
license to the patent is granted without explicit language, beyond use
in the GPL'ed code).

One may presume that the right for anyone using SL to render someone's
avatar on their client is implied in the act of putting it on SL.   This
is like the implied licenses to display copyrighted images on a web page
for people who go to that web page in Firefox, I would think.  Such
licenses are limited - you can't just copy a web page and put it into a
book without licensing the images.

So I don't get why we are thinking there is anything new between SL and
Croquet.

Other than, as I said, the idea that Croquet is being painted as a sort
of "burglary tool" that will cause all of its users to steal stuff from SL.

That's just a provocative hypothetical.   Trolling to get people angry.




Erik Anderson wrote:

David P. Reed wrote:
> Who exactly is advocating theft of property because it isn't real?  
> This seems like trolling for an opponent to me.
What I was responding to was this:

 >There's also been some concern about issues of ownership. While a
Croquet/SL interaction might highlight some issues of "intellectual
 >property", I think it's reasonable to ask: "Are these proposals
fundamentally different than changes that are happening in the world
 >anyway?"  If not, then don't worry about it.  While machinery to
protect secrets is a good and reasonable thing, I feel it's untenable to
 >permit people access to something that they are then not allowed to
use.  If the goal of a Croquet/SL bridge is to try "something fun," why
 >get bogged down limiting non-scarce, non-real property? That's as
useless as preparing to fight the previous war.

And I responded:

Because you can get sued by the people who make their real life living
creating and selling such property.

Etc..

A simple "why?" question with long-winded response.

Andreas Raab wrote:
My response about intellectual property theft in the interaction between
Croquet and Second Life was in response to this poorly thought out
remark. There are people who appear on the cover of Business Week
because they earn 6 figure incomes in Second Life buying and selling
virtual reality property after adding "non-scarce, non-real" virtual
buildings and virtual ornamentation to the "land ". That's reason and
clout enough to sue:

 >There's also been some concern about issues of ownership. While a
Croquet/SL interaction might highlight some issues of "intellectual
 >property", I think it's reasonable to ask: "Are these proposals
fundamentally different than changes that are happening in the world
 >anyway?"  If not, then don't worry about it.  While machinery to
protect secrets is a good and reasonable thing, I feel it's untenable to
 >permit people access to something that they are then not allowed to
use.  If the goal of a Croquet/SL bridge is to try "something fun," why
 >get bogged down limiting non-scarce, non-real property? That's as
useless as preparing to fight the previous war.

Erik Anderson wrote:
Absolutely. While I got long-winded and rather sarcastic, that was my
central point. *I* haven't made a single Linden Buck of the the sale of
anything in SL, but there are people who have made the cover of Business
Week because they made a million dollars American in the past 12 months
since the "Linden Exchange" opened, where people can sell Lindens for
real world cash.


There are roughly 4,000 people in Second Life who make $500 or more a
month, at least on paper, which translates to 135,000L$ per month.
 There are over 100 people in Second Life who make $5,000 or month per
month, at least on paper.

There have been a few L$1,000,000 transactions in Second Life. That's
roughly $40,000 American for the purchase of a single virtual reality
item (probably a highly developed commercial island in a good location).

Flame wars are the least of your worries if someone chooses to
"liberate" something that has a demonstrated value of $40,000 in real money.

David Gleason wrote:
Second Life is free for most people. 99 % of the accounts have likely
never paid a penny to Linden Labs. Land costs real money to maintain the
servers. Second Tier land costs real money plus rental fees to the first
tier landlords who set up the server directly with Linden Labs. Premium
customers pay a monthly fee and the land fee is waived by Linden Labs
for a plot up to 16m^2 for premium customers. If you own more land than
that you must pay more plus whatever the rent is.

Lindens can be acquired by purchasing them for real money (Premium
members get $L250 per week while free members get a one-time seed
account of L$250). You can also make items for free and sell them for
Lindens. Converting Lindens to real money also costs a fee.



The client is already Open Source and you can attend open source bug
triage meetings and so on every week in-world. Quite insightful as to
how programmers deal with the virtual world while conducting business.
There is a plan to make the server open source for every aspect but the
Havok physics engine (which could be replaced by something else in a
completely open version) and eventually to allow Linden Labs servers to
communicate transparently with independently operated servers.

David P. Reed wrote:
As I already pointed out, while I may have overreacted, I don't think I
was the one doing the trolling.

Lawson English wrote:
>
>
> Flame wars are the least of your worries if someone chooses to
> "liberate" something that has a demonstrated value of $40,000 in real
> money.
>

I'm off by a factor of 10 there. L$1,000,000 is US$4,000, not US$40,000.

However, last month there were ~250 transactions worth at last
L$500,000, or at least US$2,000. Still high enough to get you a
grand-theft charge, and not just sued for copyright violations.


http://secondlife.com/whatis/economy_stats.php

The CopyBot authors were not stunned at all. I read the widely publicized chat transcripts where the coders responsible basically mocked IP rights of the silly non-coders in SL who spend time creating such trivial things as virtual hair and shoes. The people who authored CopyBot had very little respect for the content their scripts copied. And I had little respect for their "work" given that it's SO much easier to copy static content than it would be for them to replicate the *real SL objects* (meaning the entire object including its permission flags). The script-kiddie mentality is to do things that are easy to do, even if they are malicious in effect, for the thrill of seeing them done. When porting "identity" from SL to other platforms I really hope people take the mentality of an engineer rather than a script-kiddie.

Also, I hope that people come to understand identity as more than appearance in virtual worlds. In TCP/IP transactions, identity is not presumed to be the source IP since the appearance of such info is completely spoofable. Croquet security is a murky thing (I have yet to understand how permissions will be handled in a private, mission-critical croquet space and the only secure croquet application, Qwaq Forum, is too expensive for my blood to learn how it works). The username and the visual adornments of an avatar are not "identity" unless there is a system that prevents another agent from assuming the same identity and affecting the reputation or privileges associated with that identity. As SL stands, two avatars can look precisely identical. Only the username is unique in the system. To port the username as a signifier of SL identity in Croquet would mean using the SL server to verify that the user is logged into SL as that username (not just a Croquet savvy user spoofing the SL-to-Croquet data). Relying on SL as the verification has some major implications for "trusting" identity to many proprietary systems, hence I wondered if OpenID was a better solution (with SL ideally supporting OpenID rather than the SL login server becoming the ad hoc 3rd party ID system for other 3D worlds, most of which will have their own proprietary ID system as well). Really the whole 3D world platform industry could benefit from an external system like OpenID instead of giving any one world that kind of influence and central role in issues of interoperability.

Howard very clearly stated the following:
> If the goal of a Croquet/SL bridge is to try "something
>fun," why get bogged down limiting non-scarce, non-real property?

I was shocked by this because I read Howard's blog and considered him to be hip to the factors that make virtual collaboration spaces valuable. Surely Howard wouldn't describe the proprietary solutions created for Qwaq forums as "non-scarce, non-real property" but there is NO essential difference between a block of code and a block of characters that translate to 3D mesh definitions or 2D bitmaps in certain programs. To advocate straight-up copying of proprietary code (not even reverse-engineering but simple data copying at the file level) just "to try something fun," would be fairly outrageous on most developer lists (at least to everyone making some part of their living off proprietary, closed source code). However, we just saw someone advocate just that concerning the work of non-programmers. It's a trivial matter to take static content and copy it, so what is the purpose? Why not pursue the greater challenge and reproduce a Second Life object with it's full data-set. Persist the perms system from SL into Croquet. Or is that too difficult, and is this SL-to-Croquet bridge just a one-off for some script-kiddies? You can copy the appearance of the thing, but if you can't persist any of its values, any of the behaviors or attributes that define the essence of the thing as it was envisioned and crafted by its creators... well, then you haven't really done much, in my opinion. You've done the minimum to create a Windows clipboard between two applications, but you haven't even scratched the surface of virtual world platform interoperability.

I also want to address Howard's notion of what the web does well versus Second Life and Croquet. Last I checked the web was hosting far more useful persistent data stores and multi-user applications than either SL or Croquet combined. The web is NOT simply a publishing medium and it's core strength has been the interactivity and human-to-human and application-to-application connections it affords both clients and servers. The web is about communications, and it has even explored the potential of 3D avatar-based communications. Let's not forget that while 3D web failed because 3D itself offered very little value above and beyond simpler, more accessible 2D applications and interactive forums, the 2D web thrives at providing many of the activities that some people define as "strengths" of virtual world platforms. Be very careful in your analysis of these different online communication platforms that you look at the real world facts, not just the big-picture vision statements for different technologies. Second Life and Croquet were intended to reach a lot more people with a lot more value than it currently has provided, but it is the Web (with it's open, accessible standards) that is currently hosting the interactive spaces where billions of people go for information, education, research, commerce, banking, business, self-expression, identity-construction, socialization, and collaboration. The web provides the front-end, the lobby, the gateway interface to nearly every application online (only the techies know how to get to content online without cracking open a browser).

Let us not forget that the web is what currently delivers Croquet and Second Life unto its users, not vice versa. The web is extraordinarily effective at connecting people and information, and its strength lies not in its "ease of publishing" (since there is nothing all that easy about delivering enterprise level web applications), but its strength seems to be it's open standards and relative ease of use for END USERS not content publishers. In the early to mid nineties, it was not considered easy for people to create their own web page. WYSIWYG editors were all the rage because the average person lacked the code savvy to edit a few lines of html mark-up. Publishing web sites was not easy until the tools to make them (and public web literacy) were ubiquitous, but using the web (using a web browser) was very easy. It's strength was providing end users simple, easily mapped interfaces (a link is a link, things that look like buttons usually act like links, an empty box with a blinky cursor means you can type in it, an arrow button next to the empty box means there's a list of options for stuff you can type in the box, Submit means you're done and sending the form, etc.). 3D spaces are not so easy, in fact studies have shown (in gaming and academia) that some people are utterly disoriented trying to navigate a 3D space, especially one that is highly dynamic and symbolic (we are not used to 3D spaces with completely mutable landscapes and landmarks). The web has strengths that can teach makers of 3D spaces about what works, what is successful, for communications and information navigation online. It seems to me that all of the 3D platforms need to pay more attention to why the 3D web failed to gain traction with users. Trumpeting the immersive value of 3D before it's proven itself a fraction as useful as 2D online apps is a little strange. Croquet and SL may be the future of communications and collaboration online, but right now the world wide web is king of those domains.

Linking SL to Croquet is like linking two mom-n-pop drug store chains together in the face of hundreds of neighboring Wal-Marts (you might retain the people who can only get their one-of-a-kinds from your shop but the masses will go for their daily essentials to the vendor offering the most value, convenience, and accessibility). I'm trying to say that these 3D platforms are positively dwarfed by the success and market penetration of the web. And Croquet is in direct competition with the web and SL. As an online application developer, I can compare all three platforms. The creators of Croquet may perceive Croquet as having so many unique selling points as to be in a league of its own, but everyday developers and users will only be looking at the current accomplishments (live applications, market penetration, usability, etc.) rather than the promise of what might be developed on a platform in the future. SL has some really interesting content (as people on this thread have highlighted) but where they fail is clearly in the realm of usability (less than 10% adoption rate is NOT cool for something supposed to be the "future" of online applications and real-time communications). So we can link the 3D world platforms up all we want but at some point, one of them will realize that the real bridge needed is the one from the 2D web to ANY of the 3D worlds. Just as the web needed the metaphor of the printed word (web pages are "pages" for a reason) to bring users comfortably into the online web, 3D worlds need to transition users thusly, something that the games industry is just learning itself (they wonder why the most popular, widely used online games are not 3D MMORPGs but are actually web-based 2D games).

And no, just because Coke-a-Cola and soccer has more active 'users' than SL doesn't mean I advocate Croquet doing more integration work with them... the world wide web is not just any arbitrary user group that randomly could migrate to Croquet (as developers and users). The web is the golden standard for useful, accessible online interactions, and the new 3D platforms ignore its dominance and its mistakes/missteps at their own peril. And not to poke a raw nerve but... open source web projects have little trouble attracting developers. There are far more people willing to develop on a platform that has a healthy heap of end-users than a platform that looks cool and gets their geek senses tingling (believe me that croquet is a dev turn on-- part of why I'm so interested in croquet is NOT that it is useful to me as a technology right now but that it excites me with its innovations and promise for the future). A web lobby for croquet would be a much more attractive project for both devs and users than a bridge between SL and Croquet, although yes, SL has the most users of the two platforms. That people inside SL want to bridge the two is fine (so long as they don't create a tool to mass-transfer assets they have no right to transfer: objects they "own" but which are no copy/no mod should not be copied to Croquet imo). But my point remains that no one should get their hopes up that this bridge will be a huge boon to Croquet (I once thought that myself! But then I considered further, with the insight of someone who actually has been an SL resident for many years... quite different than many in this conversation who only analyze SL from the outside looking in). SL is very cool (and I can't stress enough that I want them to succeed and fix things that are not working in SL). But Croquet offers a unique value that has more in common, philosophically and technologically, with the world wide web than Croquet has with SL and its proprietary playground which LL graciously leases to developers. Croquet is the real deal: an open platform with open values. I see no better, larger, or more like-minded ally for Croquet than web developers. In fact, I'm so stoked on this point that I will bow out of this SL thread (sorry for hijacking it in the first place with these web-related ideas). And I'll start a new thread on proposing a web based lobby. Surely the SL bridge will happen no matter what, but I also really want to see the web bridge and I can see now that it's best proposed as a separate issue entirely.

If you're still skeptical of the value of the web in terms of virtual worlds, consider just this one tiny aspect of Second Life: how much more usable would SL be if they had not implemented their own search feature? What if they had tightly integrated their data with established web search? Think of it from a user perspective: choosing Google or Altavista from a drop down (ala your web browswer) to spider SL-tagged xml content would yield much better results in-world and the bonus would have been the natural broadcast (via metadata) of ALL SL content/spaces across web search engines (thus searching for a humanism group from google on your desktop would bring up not just the humanist groups on the web but the ones *inside* SL). But nooooo, SL was created with too much proprietary crap and now they will be playing catch up trying to bridge their walled garden to the entire rest of the online world.

Anyways, I recognize that the value of web integration is far off topic from the implementation issues of an SL-to-Croquet bridge but thank you for reading so far and please add your comments about this on the new thread I'll start just to kick around the idea of a web front end for Croquet (possibly using Seaside, a web application framework in Squeak).

Cheers,
Kelly

On Jul 26, 2007, at 5:52 PM, Lawson English wrote:

It seems to me that you've been quite imprecise with your pronouns  
throughout this discussion; this might be the cause of some of the  
friction. If someone else chooses to violate a right/powerful  
entity's copyright, then they might find themselves with bigger  
problems than a flame war, sure.  As a Croquet developer who has not  
violated anyone's copyright, I'm not worried; I don't consider it any  
more probable that I would be sued than that someone would sue the  
Apache developers due to the existence of a pirate video site.

Howard said:  "While machinery to protect secrets is a good and  
reasonable thing, I feel it's untenable to permit people access to  
something that they are then not allowed to use."

Rereading it several times, I cannot see an incitement to copyright  
violation.  Maybe it's just because I know him, but I think it's a  
case of you projecting your own hang-ups onto his words.  Speaking  
precisely, "untenable" means "being such that defense or maintenance  
is impossible".  In my opinion, this is a reasonable characterization  
of the technical difficulties associated with disallowing copyright  
violations in an open-source virtual world.  He's not saying "It's  
free!  Let's help ourselves!!!".  He's saying that it's an unwise use  
of our scarce resources to try to divert the flow of this particular  
river.  I can't disagree.  Others might, although preferably with a  
rational argument rather than inflammatory accusations ;-)

Josh

On Jul 26, 2007, at 6:33 PM, Lawson English wrote:

Oh boy, can we please stop with the "piracy is theft" fallacy?  In  
the USA, the RIAA lobbyists haven't successfully criminalized  
copyright violation (and it probably wouldn't be in their financial  
interest to do so); it remains a civil matter.  There are plenty of  
kids who've been caught red-handed with terabytes of movies and music  
and although many have been successfully sued, I have heard of zero  
who have ended up behind bars.

Please do not understand by this that I am condoning piracy (in  
another email, Kelly correctly highlights the hypocrisy of doing  
so).  It's just that it's difficult to have a rational discussion  
(and to maintain my usual cheery disposition :-) with someone who  
keeps repeating the same fallacies, all the while claiming that  
others are trolling them.

The original post seems to be a sincere effort to kindle a mutually-
beneficial partnership between Croquet and Second Life.  Let's try to  
get back on that vibe, shall we?

Josh

On Jul 26, 2007, at 7:36 PM, Kelly Rued wrote:

OpenID looks quite useful.  Do you have any idea how much uptake they  
already have?

As I said elsewhere, I'm not sure that Howard did advocate that;  
that's not how I read it, anyway.

Let's say that we were serious about protecting content.  How could  
we possibly do it?  Any moderately skilled hacker could modify the  
open-source client to extract the content.  Perhaps (waving hands  
wildly) it would be possible by hooking into platform-specific APIs  
to DRM hardware, but the smarter black-hats would certainly be better  
at it than we are (since it's their speciality), and only one has to  
let the cat out of the bag.  Personally, I'm not at all inclined to  
waste my time.

> It's a trivial matter to take static content and copy it, so what  
> is the purpose? Why not pursue the greater challenge and reproduce  
> a Second Life object with it's full data-set.

Because, why embark on a project that would be destined to converge  
on the lowest common denominator (whichever denominator that turns  
out to be ;-) ).

I think it's very useful to unify notions of identity, and to  
seamlessly make transitions between SL and Croquet worlds.  But  
making SL objects able to exist in Croquet spaces and vice-versa  
doesn't seem to be a very good idea at all, not when we still have so  
much to learn about the "right" way to do things.

<snip bunch of stuff about Croquet/Web integration>

I don't have time (or maybe even constructive to add) to the rest of  
you post, dealing with reasons that Croquet should focus on web  
integration instead of Second Life integration.  Very insightful,  
thanks.

Josh

> There have been a few L$1,000,000 transactions in Second Life. That's
> roughly $40,000 American for the purchase of a single virtual reality
> item (probably a highly developed commercial island in a good
location).
>
> Flame wars are the least of your worries if someone chooses to
> "liberate" something that has a demonstrated value of $40,000 in real
money.

Doesn't the monetary value of a 'commercial island' in a 'good
location' depend entirely on it being present in Second Life? If data
is ripped and transferred into Croquet, the quality of the location
that the object occupied in Second Life become meaningless, and
therefore the value, which is dependant on the location, also becomes
irrelevant.

The value is linked not only to the data, but the context in which it
is used. Just because something is worth $4,000 in SL doesn't mean it
will automatically be 'worth' that much in another environment. This is
as true in real life as it is in virtual worlds.

Phil

Kelly Rued wrote: I should point out that my original proposal had wording something along
the lines of "with some sub-set of the SL avatar cloned into Croquet..."

I deliberately didn't define what that meant because I wasn't sure  what
was practical in a quick and dirty sense, and I certainly had no intent
of anything that existed with "no copy/no transfer" permissions to be
included in that quick and dirty transfer anyway.

The idea was simply to allow *something* of the SL environment and
avatar to emerge in Croquet so that there was a feel of transition
rather than simply starting a new application. If all that can survive
due to practical and copyrighted purposes is the name, than so be it.
Certainly, my first pass will only use names. SL users of Croquet can
certainly verify if the person behind avatar they are looking at in
Croquet is the same as the person behind the Second LIfe avatar of the
same name by sending a private IM in SL and seeing who responds.

Of course, since I'll be adding a couple of lines of code in the SL
client to reroute the SL IM in and out of Croquet, this response could
be spoofed, but I have no intention of writing "spoofing" of character
IM into that or any other mod, so MY distribution won't be
spoof-enabled. It's open source so anyone who cares to examine the code
can verify this.

In these terms, any custom SL client could be considered a copybot.

The problem seems to be with SL and not Croquet which makes content more /usable/.
___

Any custom SL client probably doesn't copy the behavior of the object which is server provided.

Croquet wouldn't help make copying an object's behavior any easier. Though, it would may creating rich behavior linked to the web and real life easier.

Darius