Giles wrote: > > P. Flavin wrote: > > >Alan & Justin: she origianally asked for the "best" loaders, > > > > and you got her to use ones that you are saying don't work ? > > > > ( see below ) > > > > People want stuff that works, not your problems. > > > > Be honest with her, and the others, please. ...
Oh great. This idiot is back. You did have to reply to him. Oh well, at least the email filters are working well. 6 months without any Flavour on this list has made it a bit dry and tasteless. Now, let's deconstruct this argument a little. Please define "best" Paul. Would you prefer to have an API that supports most of the spec, does it in a completely compliant way, supports the latest generation in development and is *completely* extensible or one that hasn't been maintained for well over 12 months, is not spec compliant and doesn't even load moderately complex files? So you say the version 1.0 code could load H-anim. Really, that's news to me. How does the Javascript work Paul? Hmmm... looking through your site, there's Nancy - but that's not Nancy, where has half the file gone Paul? My copy here, the reference one from the h-anim site is 7236 lines, yours is 1999. Mine has Javascript and the ability to animate the file in the VRML, in whatever browser I happen to choose. Yours is a mute figure that relies on having Java3D and a complete, proprietary application to animate it (and even then about 20% of your file is commented out so that the loader could even load it without crashing in the first place). You wouldn't be lying to us now Paul would you?... No, I didn't think so.... Xj3D is not complete, I'm happy to admit that. Does it implement all of the VRML97 spec? No, but we're close. Did the old code implement the spec - not bloody likely. Try doing a little scripting - show me how loadURL or createVrmlFromUrl or add/removeRoute or any other scripting API was actually implemented according to the specification. Show me code that allowed a loader to show content in more than one window simultaneously. Show me an event model that worked according to spec (or even worked for the majority of code). Where is the EAI, or support for Javascript. _Anyone_ can write a VRML loader that just loads geometry. The old code didn't even parse some files that have more esoteric structures according to the BNF (a proto/externproto that had node declarations in the definition). With the exception of externprotos and Javascript, we do all of these in a spec compliant fashion - including the EAI (about 50% implemented). I even have Javascript implementation running here, although not committed to CVS yet. So Paul, which one of your lies are you going to issue an appology for? Will it be for the loader code that you have had to modify so that it would work (VRML97 doesn't have Nurbs, how did you load those that your site proclaims you did). Will it be for the Gzip support? Will it be for the custom hacking to do any animation at all because the VRML event model implementation didn't work? Will it be for taking Sun's code and turning it into a proprietary application and not contributing the code back to the community, while you crow about it "like it's amazing, it can do anything". -- Justin Couch http://www.vlc.com.au/~justin/ Freelance Java Consultant http://www.yumetech.com/ Author, Java 3D FAQ Maintainer http://www.j3d.org/ ------------------------------------------------------------------- "Humanism is dead. Animals think, feel; so do machines now. Neither man nor woman is the measure of all things. Every organism processes data according to its domain, its environment; you, with all your brains, would be useless in a mouse's universe..." - Greg Bear, Slant ------------------------------------------------------------------- =========================================================================== To unsubscribe, send email to [EMAIL PROTECTED] and include in the body of the message "signoff JAVA3D-INTEREST". For general help, send email to [EMAIL PROTECTED] and include in the body of the message "help".
