On Tue, 29 Feb 2000, Bullard, Claude L (Len) wrote:
> SVG has a lot of support behind it beyond Adobe. I mentioned
> Adobe because I found that plug in. My guess is that this will
> replace VML (Vector Markup Language) in the MS catalog. On the
> XML list, SVG is being trumpeted as "the real killer app" and
> the app to "revolutionize the web". Seems we've heard all of
> that before. :-) Yet this is not coming from Adobe; these remarks
> are coming from folks like Tim Bray, Dave Winer, Peter Murray-Rust
> etc, who are stellars in the XML cosmos.
Yes. The Web folk (and probably the SGML folk) have been dying for a
standard content type beyond text and inline bitmaps. Here's one that
plays well with text & layout engines, and whoop-bang! There it is!
> I don't think it is killer
> but that is because I don't think "killer" makes sense anymore.
> What we need are tools that fit together easily and cheaply for
> building multimedia apps and this will definitely fit into the toolkit
> for DHTML very smoothly.
True. The first product I remember being called a "killer app" was Lotus
1-2-3. Lotus 1-2-3 wasn't a killer app because it perpetuated the
spreadsheet metaphore (already seen in VisiCalc) but because it worked on
a platform most small business and personal accountants could afford and
use (thus its wide audience) and, 1-2-3 being a spreadsheet, wasn't an app
in itself, it was an app-maker, a metaapplication that allowed its users
to create the apps they wanted, such as My Budget App and My Tax Form App.
MS attempts to do this with their office apps, one of the reasons they so
totally dominate the market today (early access to OS hooks also helps).
The web browser and HTTP infrastructure became a killer app because it
allowed for easy publishing of data. Luckily, that's what it was meant to
do. Unfortunately, it wasn't meant to be very interactive. VRML 2.0 was,
and suddenly there is a problem: the platform app wasn't intended for the
content app. VRML 2.0/97 is a great metaapp for interaction but it's not
running on the right platform in a web browser.
So for those who need to do the
> semi-complex apps with the DHTML and animation integrated
> (eg, intelligent parts libraries a la Ford, Chrysler, etc), this is
> a likely winner.
Indeed, the tools that provide for these real killer apps will succeed.
VRML with DHTML interface <-> X3D w/DOM: 2 years too late?
> Opening it up and peering, it has a lot in common with VRML
> where VRML uses primitives. The scripting interface is easier
> than what we will have with the wrapper tags in X3D but the
> download is larger than what a Core X3D should be.
The download in question is larger than what SVG should be (it also
includes the incredible overhead of a Windows code installer -- usually
adds about 500K for adding the equivilent of 150 lines of tcl/tk
scripting). But, pull out the features like antialiasing, rely on the
applet hosting code and runtime model of Java, and reduce scripting to a
Java class interface (ala Shout3D and Blaxxun3D) and an "SVG-lite" should
compare favorable with a core X3D.
What's really exciting though is the possibility to leverage SVG into
SVG+3D sometime in the future. Will that have anything to do with X3D?
<sigh> I wonder, since X3D took off in the VRML-compatible direction a
year ago instead of persuing an SVG syntax (which would have made
conversion of content more difficult -- but not perhaps impossible!).
> I'd hazard
> a guess and say this will pick up a lot of the market that core
> wanted (the ad apps) given the demos. As animation junkies,
> it will be good to be familiar with it since most of the concepts
> are easy to learn if you have a background in VRML.
I'm seeing Flash banners quite a bit these days. Since Flash is part of
the basic install of newer browsers, it's got quite a market headstart.
Most important thing for SVG is to get a good SVG-integrated XHTML browser
out there, and we should see Flash content transferring to SVG. However
slow or fast, it will take a market shift for this to happen. It will be
great for webtop boxes. Compare with MPEG-4; Blendo and similar platforms;
Interactive TV initiatives that won't die. Interesting times.
3D will be making its way until it becomes a killer tool, and along the
way stops trying to be a killer app.
JEM
[delurk]
