Ethan Zuckerman’s online home, since 2003

Why not play with Croquet?

One of the questions that appears to be on everyone’s mind at this gathering, though not always articulated, is the question of whether open standards will emerge to allow interaction and compatibility between immersive web spaces. It’s pretty central to most visions of an integrated, central metaverse… but I haven’t seen a lot of conversations about whether Second Life, for instance, will release an API…

Watching a demo last night (in the hotel bar) of Open Croquet, the obvious question sprang to my mind: why isn’t the Immersive Web world flocking towards this platform?

Open Croquet is a platform jointly developed by some luminaries of the computer science world, including Alan Kay, David Reed, David Smith and Julian Lombardi. It’s been under discussion since 1990 and under development for at least four years. I saw a demo given by Kay at least three years ago, and went home to download the platform – it was awkward and clunky to the point of unusability and I couldn’t convince even my alpha-geek friends to try it out with me.

The platform has clearly made major leaps between now and then. The demo David Smith and Julian Lombardi gave last night had jaws dropping throughout the bar. Rather than a space you visit on a server through a client, OC is a three-D operating system that allows you to create objects and spaces that can appear on your system, or be imported into other systems. Each client is a server – there are no centralized servers. Each set of interactions between clients involves a negotiation – “Okay, you want to send me three-D models, textures, some text and an audio stream. I’m on a cellphone and don’t have a 3d rendering engine, so I’ll take the audio stream and the text…”

The architecture, the rendering engine, the object model and the scripting language – which is a subset of SmallTalk, the language the system is written in – are beautifully well thought out and look great. But the system has basically no users, in part because it’s more designed as a workspace replacement than as a community space, in part because the OpenCroquet folks haven’t actually bothered recruiting any users. (As fond as I am of David Reed and David Smith, I can’t really see these guys doing the aggresively welcoming customer service the Linden Labs folks are justifiably famous for…)

It seems obvious to me that one or more of the startups in the room who are planning on building a platform for virtual worlds might want to consider becoming toolbuilders and community managers on top of the Open Croquet tool – rather than developing a rendering engine, object model and scripting language, they could use the OC tools and do the hard work of creating compelling worlds on top of this architecture.

It also strikes me that there’s a compelling business to be made importing objects from Second Life into Open Croquet. Even if OC doesn’t end up gaining a community like SL’s, OC could be used as an offline editor to build and maintain objects, and would allow those objects permanence outside of the SL world… should Linden radically change terms of service, or go out of business, the objects and the scripts on them could still exist on an open platform. (I suspect there are more than a few legal issues that arise in this scenario…)

It’s now 1pm on the last day of the conference, and the main hall is empty, as nearly every attendee is watching an impromptu demo by David and Julian – I’ll be very interested to see how this changes or shapes anyone’s thinking about open platforms for the immersive web.

6 Responses to “Why not play with Croquet?”

  1. Devil's Advocate says:

    “Watching a demo last night (in the hotel bar) of Open Croquet, the obvious question sprang to my mind: why isn’t the Immersive Web world flocking towards this platform?”

    Because its written in Smalltalk and Smalltalk is dead – right? Isn’t it? It “failed” because Java “killed” it. Or Ruby, or Python or something.

    You’ve hit upon the underlying stupidity of the computing world – great technologies are never revisited – once tried they are abandoned for the “next big thing” regardless of whether they were adequate or not.

    In Smalltalk’s case, there has never been a programming environment created to rival it in elegance, flexibility, or power, and its a whole quarter of a century old for crying out loud.

    Seriously, I hope Croquet will relaunch Smalltalk into the public eye, but there has been so much mindrot wrought by the J-heads and misguided C++ fans that I’m not holding out too much hope.

  2. mtl3p says:

    You know that I raised this idea a while ago. Here’s one response I got which I never followed up on. Do you have any info if licensing issues are one of the things holding up Croquet adoption? Or is it purely the lack of evangelists?

    “One key issue with Croquet which is probably stopping it from being more widely known and supported by free software developers is that the license for the underlying Squeak software is insufficiently free to allow inclusion in key Linux distributions such as Debian and Ubuntu. Getting that license fixed will be difficult and requires getting the likes of Apple and Disney to change the license. A blog post touching on this topic is at http://www.eschoolnews.com/eti/2005/09/001173.php

    An alternative open source immersive (but no 3d yet) peer to peer environment is Solipsis which can be found at http://solipsis.netofpeers.net/wiki2/index.php/Main_Page
    It doesn’t have the same sort of brain trust working on it as Croquet does, OTOH it is not encumbered by the same sort of licensing issues faced by Croquet.

    Recommendations: support efforts to fix the Squeak License and support Solipsis by participating in its community as users and developers. If the Solipsis community gets big enough, sooner or later someone will add 3d capabilities to it. One interesting possibility is to build both a WorldForge client and server into the Solipsis node.”

  3. Ethan says:

    Smalltalk certainly gets a bad rap, Devil’sA… from me as well,sometimes. I admire the elegance, but find that it’s not generally the first tool I’d reach for to solve a problem… But I’d love an excuse to learn it better and really understand it. If OC gets some adherents, perhaps that will happen.

    mlt3p, I didn’t hear anyone talking licensing issues here… then again, two of the big players here are SecondLife and Multiverse, both of which have licensing schemes pretty far from what you’ve proposed. I don’t know that you’d have much success getting Alan Kay away from the Squeak license, but I can put that question to David Smith, if you’d like…

  4. The Croquet license is not the same as the Squeak license. Instead, it is basically the simple MIT license. Do whatever you want with it with NO restrictions on the code. The only restriction is that you may not call your system Croquet if it does not interoperate with the version provided via Viewpoints Research – Alan Kay’s non-profit, and available on the Open Croquet web site. They own the trademark. This is the way we intend to insure we don’t have fragmentation of the community, but still allow people to do what they want.

  5. mtl3p says:

    cool. thanks for the info.

  6. peter says:

    It’s great that Croquet uses the MIT license; but doesn’t the fact that it’s built in Squeak, which does not use the MIT license, make that almost irrelevant?

Trackbacks/Pingbacks

  1. Raph’s Website » Metaverse Roadmap roundup - [...] Ethan’s report on OpenCroquet The demo David Smith and Julian Lombardi gave last night had jaws dropping throughout the …
  2. Wetmachine » Inventing the Future » News from the Metaverse - [...] Good blog [...]

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

 

Powered by WordPress | Designed by Elegant Themes