- From: Vincent Hardy <vincent.hardy@gmail.com>
- Date: Wed, 2 Nov 2011 08:48:31 -0700
- To: Kristian Sons <kristian.sons@dfki.de>
- Cc: "public-declarative3d@w3.org" <public-declarative3d@w3.org>, "Ntrevett@nvidia.com" <Ntrevett@nvidia.com>
Hi Kristian, (please forward to public-declarative3d@w3.org if needed since I am not a member yet) Thanks for adding me to the loop. One thing I wanted to bring your attention to is the 'transform-style' property in the CSS 3D transform spec: http://dev.w3.org/csswg/css3-3d-transforms/#transform-style-property Kind regards, Vincent On Nov 2, 2011, at 6:54 AM, Kristian Sons wrote: > [added Vincent and Neil in CC] > > Thanks Don for taking the minutes of our CG meeting yesterday. > > Let me go further in detail about the discussion we had with Vincent: > > Vincent raised the question, why not to transform the HTML rendering > from discreet layering to a real spatial rendering. We had this > discussion during Web3D 2009 in Darmstadt, but gave up this idea quickly > because it seemed to far away from anything doable in mid-term. > > But with the transformation of compositing from software to GPU, > performing a orthogonal projection of the layers in 3D space, the idea > appears no longer that bizarre. > > I don't want to say: Let's force this evolution. It's hard enough for us > to find (browser vendor) friends with the idea to include a 3D rendering > engine into the Browser :) But this silent evolution might play into our > hands as the HTML rendering engine includes more and more 3D techniques. > > Just imagine what a real depth system could bring: > - 3D Objects could cast shadows on a web page (or shadows between layers) > - Objects could go through a web page > - A real lighting system also for 2D > - Depth of field > > The Tilt project goes a little in this direction. Still discreet > one-per-object depths, but it's easy to imagine to break that. > > Sure, there are many problems to be solved, especially perspective > projection will not help reading text. > > Coming back from these visions, we also had some very concrete > discussions. Vincent told us about efforts to merge the CSS 3D > transforms and transformations in SVG which will also influence the used > Base Types (CSSMatrix, SVGMatrix). Vicent said it would be nice if we > participate and bring in our knowledge and requirements. > > This seems to be a promising start to coordinate our efforts with > efforts in current W3C WGs. > > Best regards, > Kristian > > > Am 01.11.2011 23:38, schrieb Don Brutzman: >> added links, now that networking is available >> >> On 11/1/2011 1:52 PM, Don Brutzman wrote: >>> [...] >>> 2. We asked about the W3C AR community group. There was no meeting >>> for this group at TPAC, and the mailing list has been silent for >>> about 2 months. Not sure if they are still active. >> >> http://www.w3.org/community/ar/ >> >> no wiki page found >> >>> Discussion question: shouldn't this group have an AR use case? >>> We all agreed this was valuable to express as part of Declarative 3D. >>> We can distill our use cases from descriptions available on that site, >> >> none found >> >>> from the recent AR Standards Group meeting, from the new ISO SC24 >>> working group 9, and from the X3D AR working group. >> >> X3D AR working group: >> http://www.web3d.org/realtime-3d/augmented-reality-ar >> >> ISO JTC 1/SC 24 >> http://www.iso.org/iso/iso_technical_committee?commid=45252 >> slideset >> http://www.perey.com/ARStandards/%5BKim%5DISO_SC24_Status-Slides.pdf >> >>> Anita gave us a summary report of the just-completed AR Standards Group >>> 2-day meeting, 15 people attended. We discussed various standards >>> alignment possibilities. >> >> http://www.perey.com/ARStandards/ >> >>> 3. Process suggestion: declarative 3d community group teleconference >>> efforts probably ought to start using tools available to all W3C >>> working groups: >>> - IRC chat, scribing, and web logging for meeting minutes >> >> Details >> http://www.w3.org/Project/IRC/ >> >> irc://irc.w3.org:6665/#declarative3d >> or >> http://irc.w3.org/#declarative3d >> >>> - Action items list and issues list >>> >>> 4. Discussed agenda for Wednesday afternoon breakout session. Reviewed >>> and discussed the Declarative 3D Community Group slideset in complete >>> detail. Kristian and Johannes presented a large and quite-impressive >>> series of demos. >> >> http://www.xml3d.org >> >> http://www.x3dom.org >> >>> Considered current issues and past lessons learned >>> in other W3C standards. >> >> Recommended reading: Tim Berners-Lee, _Weaving the Web_ >> http://www.w3.org/People/Berners-Lee/Weaving/Overview.html >> >>> 5. Vincent provided a short summary of SVG filter effects as a way >>> to predictably offer underlying shader capabilities to SVG authors. >>> We talked at length about the effects work going on in CSS, he gave >>> a demo showing lots of impressive capabilities, and we discussed >>> some CSS/shader technical challenges where a 3D approach can help. >> >> http://www.w3.org/TR/SVG/filters.html >> >> He also mentioned CSS Regions and CSS Exclusions: >> http://dev.w3.org/csswg/css3-regions >> http://dev.w3.org/csswg/css3-exclusions >> >>> 6. Neil discussed rationale for a 3D transmission format. At some >>> point, 3D assets need to go across the network and it might be a >>> valuable addition to ecosystem to have a format for 3D analagous >>> to JPEG for imagery, MP3 for audio, H264/MPEG4 for video, etc. >>> Alternative is that browsers have to do decoding in javascript >>> or content doesn't work. Seems like an obvious hole. Could do >>> an inventory of available work, discuss possibilities with MPEG etc. >>> Progressive streaming and compression would be primary benefits. >>> >>> Interesting discussion (once again). >> >> A future teleconference will be setup to discuss further >> for those who are interested. >> >>> ====================================== >>> Here are things we didn't discuss yet. >>> >>> 7. Requests for W3C >>> - (Need to confirm) community group support for IRC chat, trackers >> >> confirmed available, links above >> >>> - Occasional team review, feedback regarding goals and activities >>> >>> 8. Areas where declarative 3D is very different from other media, >>> languages and web standards. Rationale: establishing W3C Recommendation >>> for Declarative 3D should add new capabilities to the Web. Undesired >>> alternative: W3C members unconvinced that new work adds value. >>> >>> - User-driven navigation in and around 3D models >>> - User-driven interaction: selection, picking, collision detection >>> - Reuse and composition of individual 3D models into larger worlds >>> - Full integration of 3D models with HTML5 content by Web authors, >>> allowing 2-way exchange of events via DOM >>> - Augmented Reality (AR) with user control of 3D display space, >>> not simply 2D overlays on top of video feed >>> - Embeddable physics >>> - >>> - >>> - no doubt several others... >>> >>> >>> 9. We need to characterize shared relationships (+) and distinguishing >>> differences (-) between Declarative 3D goals and other W3C technologies. >>> >>> + integratable with HTML, as external or embedded page object >>> - different than SVG, native 3D representations essential, though many >>> lessons of SVG design are relevant >>> - different than canvas: not low-level shader programming >>> + DOM representation central to capture 3D scene graph. Holding data >>> structures in real time is now satisfactory with high-performance >>> DOM implementations. >>> + DOM event model is dissimilar to typical event-passing paradigm of >>> most 3D scene graphs, but can be aligned >>> + HTML5 interaction model needs to be utilized, e.g. HTML DOM level 2 >>> mouse events >>> + Significant level of CSS capabilities can be utilized, possibly >>> align HTML colors with 3D materials >>> - Need consistent accessibility so that navigation and usability is >>> intuitive/predictable (rather than "lost is space") >>> >>> Additional potential relationships with W3C Recommendations: >>> + Most base types supported in XML Schema - add n-tuple arrays? >>> ? Multimedia synchronization - other requirements beyond HTML5? >>> ? XPath and XSLT support needed, possible? >>> ? MathML support for equations? >>> ? SVG, movie, canvas support for creating embedded texture images? >>> >>> One way to get completely confounded would be if we cherry-picked >>> pieces and parts of W3C Recommendations. Instead we should accept >>> specific requirements and interoperability to align Declarative 3D >>> compatibly with Open Web Platform. >> >> http://www.w3.org/wiki/Open_Web_Platform >> >> http://en.wikipedia.org/wiki/Open_Web_Platform >> >>> 10. Use cases and requirements. Needs work. What's the plan? >> >> http://www.w3.org/community/declarative3d/wiki/Use_Cases_and_Requirements >> http://www.w3.org/community/declarative3d/wiki/Requirement_Building_Blocks >> >> all the best, Don > > > -- > ________________________________________________________ > > Kristian Sons > Deutsches Forschungszentrum für Künstliche Intelligenz GmbH, DFKI > Agenten und Simulierte Realität > Campus, Geb. D 3 2, Raum 0.77 > 66123 Saarbrücken, Germany > > Phone: +49 681 85775-3833 > Phone: +49 681 302-3833 > Fax: +49 681 85775–2235 > kristian.sons@dfki.de > http://www.xml3d.org > > Geschäftsführung: Prof. Dr. Dr. h.c. mult. Wolfgang Wahlster (Vorsitzender) > Dr. Walter Olthoff > > Vorsitzender des Aufsichtsrats: Prof. Dr. h.c. Hans A. Aukes > Amtsgericht Kaiserslautern, HRB 2313 > ________________________________________________________
Received on Thursday, 3 November 2011 17:53:09 UTC