Review Of 3D Web Browsers

shelflife points to this very intersting article on the 3D Web browsers in Scientific American. He writes that of the 3D systems mentioned, "A Swiss company, Geonova (www.geonova.ch), seems to demonstrate best that the idea of a geography-based Web is feasible with today's PCs. Engineers there created two impressively detailed models of Switzerland--one of the entire nation with 25-meter resolution and another of two central cantons at 50-centimeter resolution. .. Text and iconic labels hovered quite legibly above towns, lakes, companies and tourist attractions; clicking on the labels opened associated Web pages. What other 3D browsers are there -- VRML plugins have been around a while -- yet they do not seem to be successful. Why is that?"

Comments

Dumb question

[wrinkledshirt]

Web-based communication is largely interpreted, as opposed to compiled. Are there any successful examples of high-performance interpreted 3D out there?

why they don't work

[Teflon Coating]

yuck, the reason why they don't work are because they're slow and don't benifit any. Even most 'regular' people don't want to surf the net by clicking through an interface that looks like a town. Even if it does look nice, the fun wears off after running around in your own little virtual town when you notice that it takes twice as long to find information.

Re:why they don't work

[_ph1ux_]

well maybe so but the thing that would be good about having the ability to run around a little town like that would be getting to know the place.

if I had the option to run through the streets of a city, and get to know where the places to go are located physically I would like it. Then when I go upto the city for a night out - I would know where most of the businesses are, and would have more options as to what to do on a night - especially if I had never been to that city before.

but I would still want the traditional style of browsing... and be able to "teleport" to the front of a business in the 3d environ.

I think that it will have larger implecations as well - people would correlate brick-and-mortar with cyberspace a bit better, as you would see where a company is (or how a city is) without having to be there.

so - in conclusion I would love to have the opportunity to travel through virtually - and what would be interesting is to run through the streets and see all the other people on that street and be able to talk to them. just like a MMORPG.

Re:why they don't work

[chiguy]

Ignoring the low-level implementation details (i.e. specification language, size, performance) which affect user experience in obvious ways, there much more basic issues when it comes to 3D as a user interface.

Lets be sure we're asking the right questions. The most important question is what task is this 3D 'technology' supposed to support.

There are three common types of interaction with system data:

1) Visualization - Seeing the data in an organized fashion
2) Manipulation - Changing the data in some way
3) Navigation - Getting from one piece of data to another
Using 3D often aids in visualization. The two main reasons to use 3D here are when: 1) there's a 3D spatial relationship between the data (molecules) or 2) an additional dimension is useful to see a pattern. Remember that 2D is a subset of 3D and having the 3rd dimension does allow one to convey more information. Perhaps the biggest 'problem' with 3D visualization is occlusion. But if you're using 3D correctly, then the reason you're using 3D is BECAUSE of the information occlusion conveys. If occlusion is bad thing for a particular visualization, then a 2D view will likely be better.

Manipulating 3D objects is often informative because you can look at the dataset from different sides. Defining the view of 3D objects is very natural for people (like when we turn a skull around in our hands). Defining which slice of the data, using a search or coordinate criteria is often harder. So 3D is often good for manipulation.

There are some very good and pretty bad things about 3D when the tasks are to visualize and manipulate data. But the nightmare comes when one attempts to navigate data. There's a reason we like TV and movies: we don't have to physically move to get the information we're interested in. And this is where 3D apps fail in supporting users' tasks. They usually force people to physically fly through the data to get to a related, or even totally unrelated, piece of information. The alternative is to 'teleport' people to the next bit of interesting information. However, as anyone who's fallen asleep in one place and woken up in another can attest to, teleportation is an extremely disorienting experience. So this is the fatal flaw of 3D displays: flying through the data is bad, but teleporting is bad.

Conclusion

So I think 3D can be useful in supporting visualization and manipulation tasks. But it is incredibly bad for navigating through data. Most 3D web projects emphasize navigation, and this is where they fail. Applications using 3D can succeed if they concentrate on what they do best, visualization and, to a lesser extent, manipulation of data. Navigation tasks should be left to the 2D world.

Vanguard

VRML.

[garcia]

even the article mentions that one of the main reasons that VRML is not in wide use is b/c of the "getting lost" factor. People just aren't accustomed to the way that VRML represents the page. It was also noted in the page that text and VRML don't mix well together.

Apparently that's why this new fangled browser is better for 3D viewing.

I myself am quite happy w/the current status of the web but who knows what joy this could bring to seat selections for concert venues, etc ;)

Why? Two reasons

[Lish]

Why aren't VRML plugins used more?

1. It's slow.
2. No real or perceived value-add from using it. Why should I switch to a 3D filesystem browser, for example, when what I have works just fine?

Make it faster and give us a good reason to use it, and it might get more widespread use.

Adobe Atmosphere is my favorite

[disc-chord]

I'm not too impressed with anything I've seen from VRML, but Adobe's Atmoshpere was a bit of an eye opener. Check it out yourself at Adobe's Site ... free to play with while it's in beta. I haven't personally developed for this yet, but I hear it's quite intuitive. As a bonus, you have site-specific chat. Now if they will just add site-specific-VoIP I will start taking a more active interest, but for now I think all of this 3d Browser stuff is a bit academic as noone has found any particualarly good use for this stuff yet (at least nothing that the mainstream surfer is going to clamour for).

Why not VRML? VRML sucks.

[RobertGraham]

The question in the article was: VRML plugins have been around a while -- yet they do not seem to be successful. Why is that?

This isn't a flame, but the answer is simply that nobody likes VRML. People think that things are successful because of some other force than people like something. This can be true in rare cases (huge marketing campaigns like Nike's can change what people want), but ultimately, if something isn't successful, then it is because people don't want it.

The real question is: why don't people like VRML? Well, load it up yourself and view 3D worlds. Now play Quake. The VRML experience is unsatisfying, but Quake is fun.

Here are some basic reasons why VRML fails to stimulate people:

• Navigation sucks. The controls were built for people who wanted to model 3D objects from the outside, they weren't built for people who wanted to navigate the intireors of dungeons. Few people wanted to look at the 3D objects, most people want to fly through objects.
• VRML worlds sucked. Because of (or causing) the navigation problems, most VRML were objects you attempted to manipulate rather than 3D worls you could fly through.
• VRML didn't grok "cyberspace". Go to old VRML design documents and read the description of how they define "cyberspace", then read William Gibson's defition (or any cyberpunk definition). The VRML group was trying to model the real 3D world and objects, trying to make the PC model reality. People don't want this -- they want the computer to do stuff that you can't do in the real world. Doing real world stuff is easier in the real world -- VRML brought nothing new that the real world didn't have to offer. (This is why Quake is fun: it isn't the real world -- I love the low-grave levels :-)
• Poor leadership. Read Mark Piesce's old writings and contrast with Linus/ESR's writings. Piesce is a petulant child compared to the maturity of Linus/ESR/Cox/etc. Emotional ranting is popular in forums like Slashdot, but leaders who behave that way hurt their projects.

The real answer is that 3D has taken over the world and become the driving force behind computers (e.g. 3D cards in computers have more gates than CPUs). The 3D market has expanded hugely fast. There are those that figured out how to catch the wave (John Carmack @ Id) and those that failed to grok what was going on (Mark Piesce w/ VRML). One of these days we'll see some interesting 3D technology added to browsers; it won't look like VRML, it might look like Quake/Doom or Flight Simulator, or it might be something completely different.

Re:Why not VRML? VRML sucks.

[dimator]

People think that things are successful because of some other force than people like something.

When I'm done parsing this, I'll read the rest of your post...

:P

Re:Why not VRML? VRML sucks.

[istartedi]

As someone a little bit closer to this issue (check webpage) this is why I think VRML failed.

1. VRML-97 is not a superset of VRML-1. There are features in VRML-1 that don't convert easily to VRML-97 so people who started with VRML-1 had to re-do a lot of stuff by hand. That discouraged a lot of the early movers.

2. The VRML-97 specification specified too many things that didn't need to be specified (like text layout, which looks crappy in VRML anyway) and initially failed to specify some things very well. There was some question about what scripting should be used at first, later Java and ECMAScript worked their way in but that leads me to...

3. It duplicated things that could be done with other things. In particular, you can do a lot of 3d with Java, and if you are going to use Java to script your VRML world anyway you might as well just do everything in Java which leads me to...

4. Crappy installed base. Really weak VRML browser shipped with IE and Netscape died before its decision could have made any impact.

5. Somewhat different computing paradigm. The VRML file contains "sensors" which trigger events that are processed by scripts. In other words, the data drives the code instead of the code driving the data. Is it a file format? A programming language? What is it? I'll tell you, introducing a different way of computing is fine, but they didn't pitch it that way, which tells me that it was more of an accident. It's always a bad sign when different ways of doing things get introduced by accident.

6. Bloated syntax. I know I'll catch it from some people for this, but I stand by it. Why was the proposal for VRML-97 called VRML-2? I'll tell you: because it has twice as many brackets and braces as VRML-1, and it doesn't really make things any easier to read.

7. Performance, performance, performance. A few months ago someone on comp.lang.vrml posted something that looked like a simple Quake level. It ran at 1 FPS on my box in a tiny little window. The same box runs Quake full screen at least 24 FPS, probably more but I can't tell and I don't care because Quake looks fine. The VRML performance problem is intractable too, because it doesn't have any standard way to do BSP or any of the other tricks that games do.

There are probably other reasons too; that's just the top of my list. Oh well, I had a lot of fun with it in the early days, and I learned a lot coding for it but it is DEAD, DEAD, DEAD. I use my VRML program mostly to create animated GIFs and for photo-shop like effects (layering translucent PNGs and taking screen shots is cool) and I keep the web page up because I hate to kill stuff. I harbor no delusions. VRML will never see mass appeal. It seems to have carved out a niche in some government and academic circles, but there is no excitement there, no profit, and not much life.

Kinda cool

[zulux]

The demo is really neet - it's worth firing up a Windows box. As you fly around, the system feeds you better and better textures for the ground. Unfortunatly, it seems that it dumps the good textrues as soon as you fly away too far. The user interface works ok - user '-' and '+' on the numeric keypad to change elevations.

3D requires interaction

[mj6798]

Computer screens are generally 2D, without even depth information. When we talk about "3D graphics", what it means is that we get a 2D image that transforms like the image of a 3D object when we interact with it (mouse, trackball, etc.). That is, 3D graphics is simply a particular way of showing and hiding information in response to user input. So, we need to evaluate it relative to other ways of interacting with data on the screen.

I think in practice it has turned out that mechanisms like a folding tree and a tabbed window are more effective and easier to use for interacting with the kinds of data we deal with. Our current 2D interfaces require less user interaction and intervention and display more useful and related information than a system that is constrained by trying to conform to the rules of 3D objects. The closest to 3D we have come in user interfaces is zoomable UIs, but even there the jury is still out whether that is actually useful in practice.

A very good question.

[tshak]

Well, it doesn't _HAVE_ to be interpreted. One reason FLASH runs so well (considering all the real-time vector that it has to do) is because it's compiled. Solutions like VRML will never work (IMHO) because of the fact that they're interpreted. (Yes, I know some of you think it's this huge hog, but my PII350 runs all but the lamest FLASH at full speed).

I can prove that VRML will never catch on...

[aiken_d]

There is no VRML porn.

'nuff said.

-b

Re:I can prove that VRML will never catch on...

[drodver]

Look 'harder'

Why VRML didn't work

[LazyDawg]

I think VRML failed because it was not as easily understood by the mass market. Anyone and their dog can set up a HTML page, and they usually do, but to do a VRML page they have to use a computing metaphor that most consumers and even geeks can't understand.

Interior Decorating.

What is it? You can't ask a geek to make you a stylish personal home on the web. Its just not feasable because they spent their lives reclusively, with clutter all over the room and with clean elegance on the paper products they make. Sure there is an elite few who can make WAD files already, but they cannot easily pass this knowledge on to the masses.

Speaking of WADs, there is the difficulty in scripting events in VRML. You need a plug-in for a static language that doesn't even offer realistic doors or interaction with monsters. Where's the fun in a 10 meg download that doesn't even offer you a gun or sword or knife? VRML came out about the same time as games which for the first time offered Deathmatch mode, so it was sadly feature-poor for its size.

If they could re-make VRML as a familiar XML-style language with some support for java and javascript, then all you would need is a good Frontpage equivalent and free interior decorating or painting courses on the web.

We've got one too!

[antarctican]

At Antarcti.ca Systems (http://antarcti.ca) we've been working on the idea of geography based navigation for over two years. Both in 2d and 3d based maps, which can be seen at our demo site of http://map.net

It's a facinating new era of data navigation, doing away with the old screens and screens of text and moving towards a paradigm closer to the real world.

Flawed Metaphor

[Greyfox]

3D has its place but every metaphor where someone has tried to make the virtual world mirror the physical one has sucked in a big way. I don't care that a web page is in Sweden or California or Canada. I just care that it has the information I want on it. Find a metaphor that organizes information in a useful fashion in 3D and I might be interested. As has been mentioned, without a 3D input device and a 3D display, it'd still be a pain to navigate it, though.

File Size, etc.

[Alien54]

I can recall just a few years a go handling a tech support call for a vrml editor product (for vrml 1.0) This poor gal was doing a fancy vrml front door as the main entry for the university library, complete with books, tables, etc with links, and walk-about navigation. At the time I spoke with her, she was editing on a winbox, and uploading to a unix server, and was running into problems because of the case-sensitivity of Unix vs Windows. Typical stuff.

She wanted this to be the main interface for the site, even for dialups. This was a time that 14.4 was still somewhat common, although they were dropping out.

She had no problems making it run quickly over the LAN, of course, to her hot top of the line box, but had lost sight of the fact of dialup speeds. Her file was 1 meg, and growing. When I informed her of the typical 14.4 speed (14.4 = roughly 100k/minute under the best conditions) she flipped. This would be ten minutes to load on a 14.4 modem. Even with a 56k modem, that would be 2.5 minutes to load.

Then you load the next page.

Yes, I know of pages with 1 meg java script files.

I think this is part of the two different design philosophies. One is to design for everyone, the most users possible. The other is to design for the power users, for the elite, the people who have the hottest software with the hottest hardware, etc.

The widest adoption requires the moving away from designing for hi end. but them you get the old arguement that if everyone has it, it is obsolete.

3d vs. 2d

[Amit J. Patel]

It's not just the web. Why do people still watch 2D movies and read 2D books? Why are so many top seller computer games 2D (SimCity, Civilization, Roller Coaster Tycoon)?

Your eyes view the world in 2D (plus depth), not full 3D. Monitors are 2D. Trying to stuff a 3D world through a 2D pipe means you lose data. For example, unless you're a hyperdimensional being, you can't see what's behind something without rearranging the data. It's "cool" but it's also a pain at times. If there's no good need/benefit for 3D, then using 3D is often worse than the 2D version. Most everything I do on the web doesn't benefit from 3D. I'd rather get the 2D version.

Read Jakob Nielsen for more thoughts on 3D vs. 2D.

- Amit

Re:3d vs. 2d

[jedwards]

Your eyes view the world in 2D (plus depth),

Or, in other words, 3D.

Come on .... if our own sense aren't 3D, then nothing is.

Re:3d vs. 2d

[jedwards]

Ehhh???? How the hell have you have you lived as long as you have if you can't see in 3D?

Don't bother answering, I don't really care ... just promise me that the next time you go for a drive in your car you phone me up and warn me first so I can get well out of your way ...

Thanks....

Re:3d vs. 2d

[jedwards]

This means that you would see the entirety of whatever you were looking at.

No it doesn't. Perceiving something in 3D means I perceive it in 3 dimensions, not from 3 dimensions. I perceive a width, a height and a depth. Count them : 3.

Re:3d vs. 2d

[jedwards]

My judging of colour is off sometimes (because of the lighting, say) does that mean I don't have truecolour vision?
My perception of width is off sometimes (optical illusion).
My perception of height is off sometimes (e.g. more optical illusions).

Damn ... I'm blind.

I don't KNOW how wide something is, I don't KNOW how tall something is and I don't KNOW how something smells or tastes or feels or sounds.

I have organs which sense depth - they're called my eyes and my brain. They've worked for me and other creatures for millions of years.

Re:3d vs. 2d

[shyster]

No, not 3d; 2 and a half, perhaps. If i saw in 3d, i'd be able to see the back of something at the same time as the front.

Ehh? WTF? I think that'd be more akin to x-ray vision...or disconnected eyeballs.

3D translates to 3 Dimensional. The dimension in question are:

1. Length
2. Width
3. Depth
If you see 1D, you'd only see lines. 2D would be able to give you shapes, but no idea how deep the shape was. 3D gives you all of it. Check out Flatland for help...which you obviously need.

Re:3d vs. 2d

[ikekrull]

Wildly inaccurate?

If we didn't perceive depth very well indeed, we wouldn't even be able to catch a ball thrown to us.

The idea that possessing another 'organ' would somehow add a new dimension to the world we sense is a little off base.

Dolphins, and bats for example, use a echo-location techniques to perceive depth, with enough accuracy to catch a fish or flying insect without any visual cues whatsoever. Our ears are just not as sensitive as theirs.

Our sense of touch is 3D in the extreme. We can feel the shape of an object, front, back and on all sides, but the nerve cells in our fingers are fundamentally no different than the nerve cells in our ear canals, its just the way we interpret the data they send, based on what we subconsciously understand about the relationship of those nerves to the rest of our world.

We don't 'know' anything at all. We only perceive what our senses tell us. You don't 'know' if something is blue or green, you can only infer it's colour from the way it looks to you.

If someone shines a red spotlight on a matte white object, and you look at it, you see a red object.

If we saw in 'True Colour' we would always know what colour an object was, even in the total absence of light. This is fallacious, of course, since the concept of colour depends totally on the existence of light.

Just as colour, the concept of 'depth' only exists because we have the means to sense it. You could surely improve or augment the way we sense depth, but would that change the way we perceive it?

X3D vs. VRML

[Boiotos]

The XML solution on the horizon is X3D. This has a much better chance of being a useful 3D markup language because in most cases it would be only one representation of the base data among a set of alterantives including vector graphics (SVG) and XHTML.

Vrml is a fancy mud/moo

[BrookHarty]

Even thou Vrml worlds are neat, most attempts have been to copy a mud or moo. Make buildings, claim some land for yourself, look at the pretty ads.

I want content, news, files, mp3, chat, something other than a landscape of nothing.

Id like to see a true representation of the Internet, nodes, routers, servers, etc. I want to travel along, stop at google, search, and take the paths from each search. Goto Slashdot and see other people reading/posting and maybe interact. Use Gnucleus and see the packets of data flow from my workstation and back out onto the network.

I want to see the Internet turned into virtual reality. Not a mud/moo in vrml.

Re:Vrml is a fancy mud/moo

[Eil]

Finally. I knew it had to be somewhere on this website, I just had to look hard enough...

Insight.

This guy has it right on about multiuser 3D worlds. It's not about the language or program, it's not about the speed of your computer. It's about what you can *do* with it. And with *all* current 3D worlds, you just can't actually do anything useful. (Minus video games, of course. I was a great Quake fanatic back in the day.)

Like most other technology, features, and hardware, video games have been there and done that before anyone else.

I especially like how the user above pictured his idea of the 3D world. My own personal idea is admittedly a bit different... Mine would be closer to that of the Metaverse in Snow Crash by Neal Stephenson. That I could walk down the street, into a coffee shop, read up on the news, and perhaps cut someone's head off over a minor dispute would get me addicted instantly.

Antartica? Geography??

[jedwards]

How many people in the world know anything about the geography of Antartica? Just the few scientists who live there ... that's who.

And what has mapping the net onto a map of Antartica got to do with anything??? I'm looking for information about carbohydrates, say;
Am I going to type "carbohydrates" into google, or am suddenly going to realise that all my answers are to be found on Filchner Island on the North coast(*) of Antartica ???

(*) and how does the geography of Antartica work anyway?? Every coast is the bloody north coast.

Re:Why? Two reasons

[the gnat]

I've seen added value. The lab I work in does some modelling of protein movements. We can use multigifs or MPEGs, but this limits the viewer to the angles we incorporate. With VRML, you can view the protein from any angle. This is something that is of interest to people in the field- I don't think standard molecular graphics packages support this type of application.

Unfortunately changing VRML standards have broken all our files, and we do not intend to replace it. All of our content is created automatically on a Linux server, ruling out pretty much any alternative (and many of our viewers will use non-Windows platforms). For a good idea what VRML can be used for, though, look at the Protein Data Bank website. Try '1tim' as the search key- click on the 'view structure' link. They've done a fabulous job with this.

Ack...

[V50]

3D Browser? For the love of god, keep that thing away from goatse.cx!!! It's bad enough as it is...

*Shudder*

Just because..

[geomcbay]

Just because something can be 3D doesn't mean it should be. This applies quite a bit to desktops and browsers.

For a lot of things a 3D interface is just too cubersome and slows you down. Why 'fly' through a modelled tunnel to follow a link in 10 seconds rather than just click some text on a web page in 1/10th of a second?

Sure, there's a place for online 3D virtual community chat -- MMORPGs proved this, as just sitting around chatting is what they do best (most don't really have much of a game there other than kill monster, get xp, kill other monster that looks slightly different, repeat forever). But trying to integrate the whole browser (or desktop) experience into 3D is insane.

Some things just work better in 2D, just like some things just work better with a CLI, just like some things work better with dedicated devices rather than general purpose computers...

Re:Just because..

[Eil]

What Neal Stephenson proposed in Snow Crash was that when you wanted to do actual computer work, you would enter your 3D office, say hello to your butler, and sit down at your virtual computer.

You don't have to actually replace your CLI or file-managing programs, just include them into the new paradigm.

Besides, who's to say that by the time these full-fledged 3D worlds are popular, that we won't have easier and more powerful ways of moving files around? I imagine that when we do, and the new tools tend to get in the way of some particular task, that they will allow us to somehow go back to the old way of doing things, just in case. Rather like how Mac OS X provides a pretty interface for all its new features, but still allows you to launch a bash shell if you need to fix something by hand.

Re:Just because..

[Chris Johnson]

http://www.airwindows.com/inventions/Depth3DDeskto p.html

There is a place for 3D on the desktop- use of the illusion of depth produced by overlapping windows, only more so. I wrote up an idea piece on that, linked above, a while ago.

Interestingly enough, MacOSX has every single bit of technical underpinning necessary to do this, in its Display PDF layer. The window-minimizing animation literally takes a full window and not only shrinks it but distorts it oddly. It would be nothing to scale windows under OSX. I'm not sure if it'd be just as easy to fade them toward a background haze color, but it's certainly eye-opening to consider. Basically, almost a year ago I was envisioning this way of using a third dimension on xterms alone, because those are already resizable in many term programs. I thought it would be impossibly tough to do that with full-on GUI apps, but OSX already does more than that- it just does not currently furnish any method for zooming specific windows forward or back in space.

Makes me wonder if it'll be possible to come up with 'hacks' on OSX's Finder and add this behavior...

I can tell you why THIS one won't succeed.

[AndyChrist]

The web is organized by TOPICs (that is, web pages tend to be devoted to one sort of information)...and while that does include geography, things that can be organized that way are only a tiny fraction of what's out there.

But people looking for a car dealership in a particular town might think...oooh, cool! Except that any 3D game is likely to be way cooler. So what's the point?

Text-based directories for local areas are easy to find, don't open up as much possibility of putting things in the wrong place, if stuff like this 3D thing were generated automagically. They're probably easier to search, too.

The reason this won't succeed, and none will for a long time, is that they aren't especially useful, cause more problems than they solve (what problem is this supposed to solve again?) and require people to go get extra software.

If we're going to see 3D web browsing in the near future, I'm betting it will be through Flash, as that is software that DOES do something relatively useful (if you like watching cartoons) and is already widely distributed.

Not everyone can use 3-D

[wadetemp]

Try making a 3-D website accessible for someone with visual or physical handicaps. It's hard enough for someone without handicaps to conceptualize and navigate 3-D spaces. I've seen many people who didn't grow up playing video games fail try to run through a complicated room in Quake... now you expect them to get content this way? For those who don't have every facility available to them these interfaces are just not useful. But, I also think we need to work on improving this... I'm not saying that 3-D navigation shouldn't become a standard eventually; but just not now. The human/computer interfaces and technologies are just not up to snuff.

"Grok"

[danarch]

OK, we're going to review the misuse of the word 'grok.'
"Grok" is not a real word. It comes from Heinlein's sci-fi book "Stranger in a Strange Land" where, conveniently, it was never really defined (because it was a "martian" word). It has commonly been defined as "getting it".

Unfortunately, "getting it" isn't very well-defined either. Most people seem to define "getting it" or "grokking it" as "agreeing with me." For instance:

VRML didn't grok "cyberspace"

VRML didn't agree with my idea of "cyberspace"

So although the "Why not VRML" post is pretty intelligent, use of the word "grok" by the post's author and a number of other 'geeks' is not. Rather, 'grok' has become merely a euphamistic way to cover the arrogance of the person who is using the word.

Woah....

[nougatmachine]

Piesce is a petulant child compared to the maturity of Linus/ ESR/Cox/etc

Dang. He must be a really interesting subject ; )

Same paradigm as HTML

[yerricde]

Somewhat different computing paradigm. The VRML file contains "sensors" which trigger events that are processed by scripts. In other words, the data drives the code instead of the code driving the data.

It's the same paradigm HTML and Doom use. An HTML file contains "links" which trigger events that are processed by scripts. A WAD file contains "lines" which trigger events that are processed by scripts.

The Sims Online

[Jagasian]

I am not sure if it is true 3D or just isometric, but eventually everyone will have their own personalized eHouse, as opposed to a webpage.

The Sims Online is basically just that, an massively multiplayer online world, where people can build their own houses, and live virtual alternate lives as criminals, playboys, doctors, etc...

You can bet The Sims Online will become the next killer app for the internet. Just as the other killer apps (chat rooms, email, instant messaging) appeal to a large audience, The Sims appealed/appeals to a large audience. Making the game massively multiplayer is obviously the next level to take the game to, and Will Wright is the man for the job.

Everquest was too geeky for the mainstream, and chat rooms are too boring... The Sims Online? Now thats going to make allot of money!

Why 3D browsing sucks

[Animats]

Navigating non-geographic information in 3D sucks.
First, there's the fact that just getting around in a 3D graphical world is hard. But more fundamentally, imposing a geographical structure on information imposes a taxonomy, like the Dewey Decimal System or the Yahoo hierarchy. If your needs don't map well to the hierarchy, searching is hard. This is why search engines and relational databases are more useful than flat indices.

Utilising 3D

[OzJuggler]

This was alluded to in "Flawed Metaphor" by Greyfox above, but wasn't quite spelled out.

The reason why 3D web browsers have not taken off is because 3D adds absolutely no value whatsoever to the activity of web browsing. To answer why, let's consider the broader question of: Why isn't interactive 3D everywhere? (ie- not just browsers)

The answer is that a spatial representation adds value only when you have spatial relationships that you can load with meaning. The meaning is part of the browsing task at hand, which means that the spatial relationships must be set-up case-by-case for each type of task that someone wants to do. One of the few situations that don't require special setup are where the task is no more than real spatial navigation - like an action game or a CFD stream function.

If you change the axes of the coordinate system from (Right,Up,Back) to (Manufacturer,Price,Item) then you now have some spatial relationships that are loaded with meaning ("higher" means "more expensive"). But you have a "space" that is no longer intuitively navigable. This is not to say that it is difficult to get around in, but that it is diffuclt to know where you are. When you find something that looks interesting, some careful thought is needed to figure out what it is that you've found in the source data. A trivial example, to be sure, but the problem becomes worse the more complicated the dataset is. (And you'd think that's where you'd need 3D the most.)

Web sites are organised by topic and if you can organise all the topics of human knowledge using only 3 spatial axes and still keep it navigable, then you've achieved what no-one else knows how to do. That is why 3D web browsing has not taken off. (There is an arbitrary relationship between physical server location and topics stored on it.)

As an aside, UI is a big issue too because it is easier for your brain to make sense of the space around it when you can physically turn your head to see everything around you. Another reason why a little monitor window in front of you is not going to blow anybody away. (Q3A excepted)

How about a 3D version of SVG?

[iabervon]

As far as I can tell, VRML was trying to be both Inventor and HTML. This was before XML got big, with people routinely using that sort of format for non-text. Inventor was a pain to actually view, and it influenced too many 3D viewers; VRML was watered down from Inventor, as well.

There's now a good example of a XML-based graphics format which is at least usable, and which will probably gain substantial browser support (even if it doesn't become popular to actually use for a while). A vector graphics format, as well, is a good basis for a 3D format, because the user will be able to change the size of objects by getting closer or farther away. Extending SVG into 3 dimensions shouldn't be too difficult; it would require 3D primitives, of course, and people would have to figure out a useful user interface, but the first is relatively trivial and the second is a universal problem.

Au Contraire

[frog51]

The Cosmo Worlds suite from SGI - you can easily knock up virtual rooms/districts/worlds in minutes with the same ease as creating HTML pages with FrontPage or similar. And it definitely has a useful place - for example:

A 3d model of an engine for mech school - click on a part and info flashes up, followed by sequence for removing/fitting the part, maybe with a virtual tutor offering advice.

You wouldn't use it for document browsing, but that isn't what it was designed to do. It really is a piece of cake to edit manually though.

The thing that astonishes me is people complaining of long download times - you can create detailed areas in a staggeringly small size if you are sensible about your image and texture maps. For really high res stuff, yes you will need more bandwidth, but that is always the case.

Why there is no 3-D /.

[Codeala]

Just because you can, doesn't mean you should. So we now have super faster computers and graphic cards that can render 1024x768 at 60fps with 32bit colours, so what? It maybe good for gaming and some very specialised research and visualisation applications, but it is no good for general web browsing.

Why? What does browsing mean? Reading (and maybe typing) text, and maybe some images here and there. So what does a 3D environment bring you in this case? Nothing. Reading an article in 3D is pointless; most people will rather stick with the good old "flat text".

The reason Virtual Reality did not catch on for web browsing is that it is not bring anything new. At least not anything people think are useful outside of some very specialised cases. For example 3D maps cool! 3D /.? No thanks.

We are still waiting for a killer (non-game) app in 3D. In the mean time, the idea of 3D browser is probably only of interest to academics researchers.

VRML == Bloat

[sjbe]

Why does no one use VRML? Because virtually no one can use VRML. Not for anything interesting or useful anyway. Ever looked at a VRML file with any significant detail to it? They are huge. Far too large to comfortably send around the web, even with a broadband connection. Even when network speed isn't a consideration, it still is too bloated to do anything useful with that can't be done much better with proprietary (and better designed) 3D model formats.

In my job we use various 3D formats daily. Various CAD system (ProE, CATIA, AutoCAD, etc) native formats, IGES, STEP, STL, and yes VRML. We do work with virtual reality technologies, 3d modelling and simulation. The only use we have ever found for VRML is as a vendor neutral format for exporting tesselated images of CAD data between a few selected pieces of software. And even then we have to be careful how big the VRML we use is. 3D Studio Max (one of the more well known apps we use) will choke on a VRML bigger than 75-100 Meg. And this is on a Dual Proc 800Mhz P3 with 2Gig of RAM! You can generate a VRML that size with one automotive assembly with a decent amount of detail to it. (a single front axle module for instance) Forget building a world, you'll have trouble building more than a few simple objects with VRML.

If we cannot use VRML for anything useful beyond simple format translation, what chance is it ever going to have to be used for something more clever? VRML is useless because it is too big and it is too slow. Simple as that. If the format sucked less, the tools would come to take advantage of it. But right now, there is no point because VRML isn't useful for anything and even with computer speed increases isn't going to be for some time to come.

If you want to see a more useful 3D tesselated format look at .JT (from EAI) Granted, it's proprietary and not without some other problems, but it is much "lighter" than VRML ever dreamed of being while still looking quite good. It's getting used for PDM (Product Data Management) systems in the business world which require a relatively light weight, CAD vendor neutral, tesselated 3D model. This market opportunity was created in part because VRML was too bloated to be useful. The opportunity is still there, but VRML in it's current form just isn't the answer.

Lots of reasons VRML failed

[workly]

1) VRML2: unimplementable.
The spec contains 3 scripting langages. There are 3 different mechanisms for animation. Everyone implemented part of it, so all VRML2 browsers were mutually incompatible, more or less, which defeated the reason to have a spec in the first place.
Blame who you like.

2) 3D content takes, literally, 2500 times longer to make than 2D content.
A business wanting to put merchandise online has to do 1 week of work per model.
A good commercial photographer can take 500 merchandise photographs a day.
Is 3D 2500 times more compelling than 2D? (No.)

3) It exists in a tricky place, half file format and half scripting language.
If you're not using motion, why not use the existing Java viewers for OBJ's?
If you're using animation, and 3d-ness isn't mission-critical, why not Flash, which is more standardized and faster?
If the point is a walkthorough, why not a sequence of pictures?
If you're trying to make a game, why use a plugin at all?

Then, the reasons others noted:
4) It's slow.
5) It's harder to navigate.
6) Everything looks crappy.
7) Mom doesn't get it.
8) You weren't allowed to sell the plugin.

Re:Rarefied Atmosphere

[FunkyChild]

Sounds like you need the (superior IMO) Blender Web Plugin. So far a beta is available for Linux and Windows, with SGI and MacOSX support on its way.

Not to mention the Blender creator software being available for Windows, Sun, Linux Alpha, Linux x86, PPC Linux, BSD x86, SGI and MacOSX soon to come. Oh and it's absolutely free.