Search this site

Partner Login

Moving Worlds - VRML2.0 a short history PDF Print E-mail

This document chronicles Mitra's recollection of the history of Moving Worlds, as it evolved into VRML2.0 and now the ISO standard VRML 97, it collects together an archive of some of the documents, and earlier versions of proposals, and may be of interest to anyone interested in the evolution of a standard.

The most important thing that I suggest noting from this chronology, is how both giving and receiving criticism led to proposals being revised to incorporate the criticism, and eventually the proposals being merged into what has become a widely accepted standard.


Of course, history always has several versions - I'll be happy to incorporate, or link to, comments or recollections that any other participant has about the process, especially if they differ from my own perspective. Chris Marrin (then at SGI) published an article on the history in WebSmith in November '96. Rodger Lea (at Sony) has a more descriptive version of the process.

I'd appreciate copies of any of the papers I'm missing - especially the early SDSC, Sony, SGI and Microsoft proposals, and critiques by those people of each other, and my proposals. Please send them to me, Mitra, at mitra at . Contacts for other key people in the process are at the bottom.

15 Jun '95 and onwards Sony published papers on their Virtual Society, covering the same areas (does anyone have the first draft)
10 July '95 San Diego Super Computer center (SDSC), published papers on Virtual Reality Behavior System (VRBS) its protocol, and on using Perl as a language for it, and a detailed protocol specification.
Aug - Sept '95 While Chief Network Architect at Worlds Inc, and a member of the 10 person standards group, the VAG, or VRML Architecture Group, I authored and published a number of papers on adding behaviors to VRML that were originally intended to become part of VRML+ but which were seperated out because they did not relate to the multi-user core of VRML+.
16 Aug 95 A proposal to support URN's, and to make the url field an MFString. The latter becme part of VRML2.0, while the former is only just (Apr '98) being integrated.
15 Sept '95 A proposal by Tom Meyer and myself called "Interfaces",  this formed the basis of a proposal by Gavin for Prototypes, which became Prototypes in VRML2.0.
21 Sept '95 A proposal for SharedState, this was a LOT simpler than Living Worlds, and if something like it had been incorported into VRML2.0 it would have simplified multi-user a lot.
21 Sept '95 A proposal for Sensors, Triggers and Connections - Similar concepts, but different concepts to, Sensors, Scripts and ROUTES.
Sept through Oct Discussions, and criticisms of each others papers led to realisations that the underlying core was common, examples of the back and forth include those below, notice how each paper critiques the parts of the other proposals not liked, and incorporates the parts liked
4 Oct 95 Sony's approach to behavior and scripting aspects of VRML: an Object-Oriented perspective
5 Oct 95 Sony comments on Interface proposal and Sensor Trigger proposal
5 Oct 95 An update of Sony's proposal for extensions to VRML1.0
9 Oct 95 A detailed critique of the SDSC, and Sony proposals and comparisom of those with my proposal, and a new version of my paper incorporating some good ideas.
20 Oct 95 I authored a first draft of a paper drawing out the similarities and differences between our proposals. By this time I had left Worlds Inc, and was in the process of forming WorldMaker.
Oct 95 Gavin Bell from SGI publishes a behavior proposal based on data-flow techniques, an early annotated version, and a slightly later version.
23 Oct '95 The VAG (which Gavin and myself are members of) produces a clarification of VRML1.0 that tidies up some outstanding issues.
24 Oct '95 The VAG produces (but does not publish) the first draft of VRML1.1
25 Oct '95 A new version of my behaviors, and API papers, incorporating good ideas from Sony and SSC
30 Oct '95 Behaviors symposium at SDSC. Sony, SDSC, and myself proposed a common paper based on the event-driven model. Here is a Report and Attendees list
1 Nov 95 Gavin of SGI writes a critique of the joint proposal of SDSC, Sony and myself.
12 Nov 95 Through November, Sony and SDSC I authored a common behaviors proposal.This proposal was largely based on defining an API between the browser and an external interpreter (of any language). several drafts including this for the API. The API didn't change, but other parts of this draft have been overwritten by later versions on 2Dec95
13 Nov '95 Critique of SGI's proposal, which is still data-flow oriented, and response to SGI's critique of an earlier version. This focusses on the differences between data-flow and events, and in particular how this effects dynamic worlds (worlds where nodes are added and deleted at run time).
14 Nov '95 Sony comments on SGI's comparisom paper
24 Nov '95 SGI is persuaded that an event-driven model is better, they re-author their paper in this form and we then start to work on integrating the approaches.
26 Nov '95 A critique of SGI's first event-driven paper.
1 Dec '95 Another version of SGI's  paper (and a diff)
1 Dec '95 A draft of a proposal for "Frames and Leaves", this proposal was key because it removed the remembered state of VRML1.0. This draft is interesting because it contains both agreed upon portions, and arguments between Gavin and myself on whether Leaf (later renamed Shape) was needed.
4 Dec '95 The last draft of the Sony,SDSC,Mitra proposal, before merging with SGI's.
4 Dec '95. A group at Microsoft circulates their own proposal, called "ActiveVRML" I don't have the original paper, but here are parts one and two of scans of a paper about AVRML, a letter, an introduction and the reference manual. and language integration, white paper,
6 Dec '95 I produce a short critique of the Microsoft proposal
11 Dec '95. A common paper is co-authored by Gavin Bell and others of SGI, myself, and Kouichi Matusda and others from Sony.
12 Dec '95 We put together a comparison of our proposal with Active Worlds.
16 Dec '95 Second draft of VRML1.1 produced, this has many of the non-behavior extensions that became  VRML2.0, including Prototyping. However this was never released, the VAG realized it would need to change from standards maker, to referee of the standards process.
VRML 95, Dec '95 The paper is presented at the VRML'95 conference in San Diego by Gavin Bell and myself.
day after VRML95, Dec '95 Microsoft holds a briefing on RBML - now renamed ActiveVRML in San Francisco, most of the key VRML people are there, and most leave more bewildered than when they arrived.
4 Jan 96 Because of the Microsoft version, the VAG puts out a RFP for anyone to submit proposals.
Jan 96 SGI devotes technical writing and other resources to cleaning up the proposal for formal submittal, Sony and myself are involved in the editing, and in bringing other companies concerns into the process.
15 Jan 96 A design document explains the "Why" of Moving Worlds and some examples.
26 Jan 96 SGI puts up a page to solicit comments - whose marketing people then claimed most of the credit!
30 Jan 96 A discussion (Mitra and Sony) and explanation of how multiuser worlds could work in Moving Worlds.
31 Jan 96 SGI produces a overview of Moving Worlds, with a summary and list of changes.
Feb 2nd, 96 The Moving Worlds paper is formally submitted for consideration for VRML2.0 by SGI, and here is the announcement and the version submitted in full and in parts (without graphics)
Feb 8, '96 All the VRML companies are invited to sign on to the Moving Worlds process - the invitation.and another version of the invitation,
Feb 12-13th 96 Press release and coverage in the Wall Street Journal,
Mar 1st '96 Sony does their own press release, correctly giving credit.
Mar '96

VRML2.0 wins the vote on www-vrml, receiving 80% of the vote against other proposals from Sun: "Holoweb" *, IBM*, Microsoft, Apple "Out of this World"*, and GMD.See the voting results.(orig)
* I would appreciate text versions of these to replace the scans.

Mar '96 Press coverage in MultiMedia World, Interactive Age, Connect
Most of 96 SGI edits the document in response to feedback.
20th May '96 SGI edits Draft 2b of the spec
24 June 96 Presentation to INET 96 (the annual conference of the internet society).
8 July 96 Sony leads the work on the Java reference for VRML2.0 including this draft 3
15th Jul 96 An updated version of the URN spec
17th July 96 Draft3 (no copy seems to exist any more) and differences from draft 2b
4th Aug, 96 Here is one version of the full spec and the changes from Draft 2b
5th Dec 96 A diagram I did of interfaces in VRML2 between different components
Dec '96

Final version published, not all issues were addressed, and I outline some key problems here.

As part of the ISO standardisation process the standard was renamed VRML97 and some cosmetic changes requested by ISO incorporated.
19th June 98 An integration of URN's into the Universal Media Framework

The only version of the final VRML95 document I can find is in Postscript, if there is an HTML version around I would appreciate it. Here is the postscript (compressed).

Contact Info for authors and others....

  • Mitra Internet Consulting
    • Mitra, now doing business consulting and development for Sustainability Businesses This e-mail address is being protected from spambots. You need JavaScript enabled to view it, or
  • Sony
    • Yasuaki Honda, working on home networking and IEEE1394 from product implentation and technical research level. This e-mail address is being protected from spambots. You need JavaScript enabled to view it
    • Kouichi Matsuda (Kou1 Ma2da), working on Community Place, development of shared 3D environments, This e-mail address is being protected from spambots. You need JavaScript enabled to view it ,
    • Rodger Lee, now heading up the operating systems lab for Sony, This e-mail address is being protected from spambots. You need JavaScript enabled to view it
  • Silicon Graphics
    • Gavin Bell - This e-mail address is being protected from spambots. You need JavaScript enabled to view it
    • Rikk Carey - This e-mail address is being protected from spambots. You need JavaScript enabled to view it
    • Chris Marin now consulting to Sony, This e-mail address is being protected from spambots. You need JavaScript enabled to view it