By Neil Schneider
We had the recent honor of interviewing Dr. Robert Cailliau, the co-developer of the World Wide Web. Jointly with Sir Tim Berners-Lee, Dr. Cailliau’s work holds a worldwide importance, and their mass appeal invention demonstrates important lessons learned along the way.
In the first interview segment, Dr. Cailliau shared some information about his career, physics, and some history about how the World Wide Web was created. In part two, we talk about the politics of the web, Dr. Cailliau’s personal experiences with stereoscopic 3D photography, and a sharp look at the relationship between the way the WWW was developed, and what ramifications this may hold for the stereoscopic 3D industry.
What was “the year of the web”? What critical elements secured the web’s success at this time?
Depends who you talk to. Some think it was 1989, when Tim produced the first proposal (that does not have the name WWW in it). Others think it was 1990, when the first server went on-line. Yet others, especially media, think it was 1993 when CERN put the technology into the public domain.
But to me it was 1994, because that was when the greatest activity took place:
— the First Conference, that brought together for the first time web developers who had until then never met,
— the conception of the Consortium that would guard the standards (the consortium was to be run by CERN and MIT, but the approval of the construction of the LHC on a tight budget made CERN get out).
— the interest of the European Commission to help WWW take off.
— the first stirrings of interest from commercial firms.
1995 was very different: I spent that on transferring the web from CERN to the INRIA-MIT consortium, and Tim had already left CERN end 1994.
Why was the W3 Consortium formed? Was the web destined to be successful before or after the consortium?
The web could have been successful without the Consortium, but it would have been limited to academia. Once there was interest from commercial software companies it was very necessary to ensure that all worked to the same standard. Whenever there is something that only works if two independently produced parts are compatible, you must work to an agreed standard.
Look at the HDDVD/BluRay disaster, or the VHS disaster. In both cases there was content served up on some medium that had to be played on a compatible player. You can either produce all content in many versions so that consumers can buy the one compatible with their player, or you can wait until the market decides which pair format “wins”. It is almost always the worst format that wins.
Netscape was trying hard to impose a set of HTML tags that would only work with the Netscape browser. Others were trying similar things. For years we had to code our pages with if-statements that would pick different code depending on the browser in which the page ended up. A complete disaster and billions of man-hours wasted.
The Consortium intended to avoid that situation but did not at first succeed. It was a long and hard battle to make all companies sit around the same table and agree on a standard.
We now have these standards: XHTML, CSS, XML, DOM, SVG. Yet Internet Explorer 7 still does not respect them well. Here too we lost about five years by companies not understanding the importance of adhering to a common standard.
Separate from the consortium, you formed the first international WWW conference. What made these conferences unique? What need were you trying to fill? Do they still run?
To my knowledge they still run. I have referred to the Conferences several times. There certainly is a need for people to get together physically in the same space for a number of days.
There are many advantages:
— you get to know each other in a different way than by e-mail or blog.
— you are away from the work environment and can concentrate on high-quality contact with like-minded people.
— you generate and absorb a lot of ideas.
— you get some credit for presenting a paper.
Today there is WiFi access in all conference rooms and that is detrimental to the contacts; I think I would disallow network access to the audience.
I ran the first one, and then we formed a legal body so that we could handle money well. There were two big conferences a year for the first two years, then once a year afterwards. I lost interest a few years ago, mainly because I got out of web technologies and things became much more specialized.
I had the intention also to gather large users, but that never happened.
I understand Tim Berners-Lee was resistant to the conferences at first. Why?
I think he thought it was a waste of time, but actually it generated a lot of synergy between developers and the Consortium. He was certainly pleased at the first one, when he saw how many people turned up and how many workshops there were.
In a Wikipedia interview, you compared the W3 Consortium and the WWW conferences to a “church” and “state” relationship. The W3 Consortium is the church, and the WWW conferences are the state. Given that you are personally atheist, I wonder if there is a deeper meaning to what you are saying. Do you think the web’s standards and future development was best served by a selection of companies and organizations, or would it have been better served by an industry that answered to its users? Should it have been a hybrid of both?
Well, given that I am indeed an atheist, I certainly want a very clean separation between church and state in the real world. However, in putting forward that analogy I just meant that the Consortium could not give ambiguous signals, it has to hold up a standard. The Conferences were to me the place where anyone could suggest any wild idea and present it to an audience without creating an impression of uncertainty about where things were going. The Consortium could look at some of the ideas and perhaps later use them.
I think that the market-economy principle (Smith’s invisible hand) is a fantastic mechanism for deciding who is the best cook or the best grocer. But it is possibly the worst mechanism to keep standards. It would almost be like letting the market decide what the value of pi was.
We have had a “market” type system to decide on units, and we have ended up with a metric system that is perfect for quick computing because it sticks to what calculators do: the decimal system. However, magnitudes are binary: the next thing after having one of something is having two of them, not ten. The Imperial system has the advantage that it recognizes this doubling/halving tendency, and it also has base 12 which allows division by 3. But no calculator, human or electronic, can deal with it! Try to find the area of a room that is 9 feet 10 inches wide and 12 feet 7 inches long.
The obviously best combination is a 12-base numbering system, then we change all our calculators to work in base 12, and then revise the metric system to match that. But how do you bring that about with a market? Never! It needs some clever people to sit down and think it through.
Microsoft, and I’m not afraid of mentioning them by name, have always tried to be the “standard” setting industry, but have they answered to their users? Definitely not.
In the domain of setting standards you need a group of intelligent people with no conflict of interest, who are focusing on making it work and keeping it open to the future. The Consortium accepts input, looks at it, makes a proposal and has it reviewed before making a definite standard. In some respects it tries to humor too many trends at once, but at least we do have workable standards.
I understand you are a fan of stereoscopic 3D (S-3D) photography. How did you get into this?
A long time ago I wanted to get depth into pictures I made. As a child I never owned a “Viewmaster”, but I used those of some friends. So I was familiar with the effect and also how it was done. I also had a picture book of the Antwerp Zoo, which was printed in blue/red with a red/blue set of spectacles. This gave black and white stereo pictures.
In 1977 or so I tried my hand at projecting 35mm slides on a metallic screen using two projectors, each equipped with a polarizing filter and spectacles for the audience with polarizing filters. This works beautifully but is complicated to set up.
Then for a long time I did nothing and relatively recently (2002) I started making some digital examples. I looked at getting a viewer with prisms, but prisms are very difficult to get. There are viewers with lenses but they are not so good.
To say I am a fan is perhaps too much. I will occasionally make some, but I have yet to find a good way of showing them without a complicated setup. One can of course put them on the screen and view them with fast shuttering spectacles, but that is difficult for a larger audience.
Ideally I would want to project onto, say, a 2.5m by 2.5m screen for an audience of about ten people and let everyone enjoy this with little or no equipment. Difficult to do if you want very high quality (1024×768 is an absolute minimum).
What equipment do you use to make 3D photos and how did you learn to use it?
Ah, I just take two shots, one after the other, moving the camera a little. If it’s close up, I move not more than the distance between one’s eyes (about 6.5 to 7 cm) but if things are far away I obtain an artificial depth perception by moving much more, sometimes a meter or so.
If you photograph a landscape from two positions a meter apart then the result looks like a miniature because when things are far away you do not really perceive stereo, you judge the distance by other means such as haziness.
When you do perceive stereo, i.e. your left eye sees different sides of objects from what your right eye sees, then it is because the objects are really close and your eyes are somewhat crossed too. Therefore we always think of stereo as close by, and the brain therefore tells you that anything in stereo must be close. This is why such landscapes look like they are miniatures watched from close by.
Tell us about some of your work. What photos are you most proud of and why?
Difficult. I really don’t specialize in stereo photos. I have a few, taken on trips, and mostly to document things I saw.
I made a whole series about the construction of the ATLAS detector at CERN and one of them was the inspiration for a souvenir that the ATLAS PR group sells to visitors. I liked that series. I have also some of plants and a few of statues on Easter Island.
I should do more, and especially I should pay more attention to making them high-quality. This interview will probably make me do just that.
What is it about stereoscopic 3D that fascinates you so much? What qualities does it add that makes it worth the extra effort and attention?
The depth is important to understand the spatial relation between components of what you see. I find it especially gratifying in close-ups of flowers. But I have not done many. I guess I’m stuck until I find a very easy way to show them.
On-screen you are limited to a distance of maybe 30, 40 cm. That implies that you cannot make the pictures larger than about 10cm on the screen. But that’s only 200 to 300 pixels! Quite bad resolution.
Yes, I think the technology to show them very easily is not available to the average user and so I’m not making many until I can show them easily – to myself in the first place.
Do you think your S-3D photography helps make the experiences more memorable?
Oh yes! But I don’t do enough of them.
The reason I named this organization “Meant to be Seen” is because people who haven’t seen S-3D in action really don’t get it. I know you aren’t a game player, but given your experience in stereoscopic 3D photography, and working on the premise that gamers have good stereoscopic 3D equipment, do you agree that when properly implemented, stereoscopic 3D technology should significantly increase the immersion and beauty found in video games? Why or why not?
Seeing things in 3D certainly adds to the aesthetic experience. I personally hate shoot-out games, but I can imagine that walking through the classic Riven or Exile labyrinths would be greatly enhanced if it were in 3D.
We should however not overdo this: the distances of objects is generally of the order of a few meters which is vastly greater than the 6-7 cm separating our eyes. That means that most objects are not seen in 3D: the images in both eyes are not sufficiently different. I don’t know at which point it becomes important, but I would suggest closer than 3m. So overdoing it by pretending that objects that are further away should be rendered in 3d would be bad and make the scene artificial, like my miniature-looking landscapes.
I hear (but again I need to see this now!) that in modern equipment, the separation can be adjusted (indeed, not too difficult to implement!) by the user. That would be a good thing to have, since it would allow contrasting a better perception of depth at the touch of a button.
I find it interesting that today’s world wide web is very much a 2D space and you are personally enamored by 3D imagery. Would you like to see a stereoscopic 3D world wide web future?
Personally I am interested in combining abstract concepts to help me understand the world around me. That usually means I need some notation: we use words and language for most everyday things, plus now also some photos and videos and even 3D. But I need also the notation of maths, chemistry and physics. The notation used there is not words or images but symbols placed in a 2D arrangement, in the form of diagrams, equations etc. There are rules for manipulating the equations.
When I look things up on the web I do not care much about any 3D effect there could be since I’m after abstract stuff. On the other hand, some understanding does need at least 3D (though how one would represent 4D or 5D I would not know). In chemistry especially one has been using 3D for a very long time: 3D molecules were the first 3D objects on the web, they date back to 1994 at least.
3D might bring great benefit in “gut” understanding by the general public, and I’m not trying to be paternalistic here. While a small number of people like me may prefer the abstract knowledge, communicating to the public at large has constraints: there is time, attention span, familiarity with maths. The latter especially is nonexistent or very low, even with many well-educated people.
I could think of web pages being laid out in 3D, each one being much like a room: the 3D would show better the parts that belong together and at the same time the overall structure of the info on that page. Links could be like doors that lead to other rooms. This would let most people understand the info faster and easier. Do not confuse this with 3D navigation in information spaces: I’m talking about a single page that expresses a single concept in many aspects and is laid out in 3D.
In 1976 Kay, Thacker and Lampson made the first 2D computer: we went from interfaces with single command lines to full 2D graphics. That was a big step in making the computer accessible to ordinary people: the graphical user interface, with overlapping windows and recognizable objects. It needed the mouse to interact with it.
Going from 2D to 3D may not be so revolutionary a step, but it will certainly make the information itself easier to grasp.
3D Cinema has become a very big deal with anywhere from 2:1 and 3:1 revenues earned versus 2D movies. Similar to the web, 3D cinema is an artistic endeavor. We are starting to see a wide selection of consumer 3D solutions for the time when 3D movies jump off to the home markets. Having gone through your experiences with the WWW, who do you think should decide on consumer stereoscopic 3D standards and why? On a similar matter, what lessons did you walk away with from the HD DVD versus BlueRay fiasco?
I already commented on the HDDVD/BluRay happening. The only time that I can remember that the industry did it right was with the music CD in 1980. All manufacturers and editors agreed to follow one single standard, well before any product was brought to market. I hope this effort will be repeated for home 3D, and I hope no one will want a replay of the BluRay battle.
The criterion is clear: if content needs to be played on a compatible device, there should be a worldwide standard for the content.
We have, by the way, another total scandal: the number of different video standards. MP4. QuickTime, AVI, WMV, … Not to speak about GSM: 900MHz, 1800MHz, 1900MHz, G3, GPRS, whatnot. That should perhaps be cleaned up first. In still images it’s all JPEG.
I think an open consortium for 3D content would be the best way to go. And I say that not because I was involved in the setting up of the W3C, but because I want to put the consumer before the shareholder. There is no point in making things incompatible on purpose.
If I said it’s the end users that made the World Wide Web the success it is today more so than the companies that set the standards, would you agree with me? Why or why not?
Certainly the authors did it and now almost anyone is an author. They began to fill the web with useful stuff when the worst of the standards battles were over. To my knowledge not one company succeeded in setting web standards, not even Microsoft.
The Consortium did manage to get its authority and that’s good. The average user does not want to know the why’s and the who’s of the standards. He/she wants to know that the page just uploaded will work on any browser. If there are variants then that puts him off or makes him choose one and ignore the rest.
Therefore I think the really great participation of the average users in the web came after the standards issue was resolved by the creation of the consortium. Firefox became the touchstone for the masses, precisely because it implements the standards well. The standards were essential to get the users in.
For the past ten plus years, the S-3D gaming industry has been in limbo because there was never a direct relationship between the video game makers and stereoscopic 3D driver developers – so there was always a compatibility problem. I formed MTBS to create a catalyst for the game developers, the S-3D manufacturers, and the end consumers to have a meeting of minds to overcome this challenge. My biggest frustration is many people are quick to sign off their empowerment to one or two corporate entities when it comes to determining industry success. When you worked with Tim Berners-Lee to develop the world wide web, you were all of four people in a CERN office with one of this century’s greatest contributions being named over a beer. Did it ever occur to you that your invention’s success would be dependent on a company like Microsoft? Why or why not?
Did it depend on Microsoft? As late as 1995 they were still thinking that Microsoft Network would wipe out the Internet. I have never felt that the web’s expansion was driven by Microsoft. On the contrary, their browsers and servers always lagged behind. I certainly have not ever used a Microsoft product to build a site nor to look at one (the exception being FrontPage for a very short-lived experiment).
For 3D there should, as always, first be a phase of wide experimentation by the pioneering enthusiasts. I feel that’s what’s happening now and that phase may be coming towards its end. The changeover from experimentation to public acceptance and consumer use can go two ways: either a single company “wins” and floods the market with its proprietary technology, or the pioneers of today act like the internet community did. Their motto is: “we don’t believe in kings, presidents or voting, we believe in rough consensus and working code”. In other words, the community itself sets the standards and above all remains deeply involved so as to keep moving forward instead of dying in the rise of a single monopoly.
This raises another point. Which browser does your personal website have problems with and why? Can you relate this challenge to the issue of stereoscopic 3D empowerment? Is it dangerous to have rogue corporate super-powers in the 3D industry? Why?
I code my pages (and I need to renovate quite a lot of them still!!) in plain and simple XHTML (strict version). Likewise for CSS. There are no problems with Safari, there are minor type-rendering problems with Firefox but there are ugly problems with Internet Explorer when it comes to positioning relative
I think any industry should watch out for rogue superpowers. So, we need a good, simple, workable standard for 3D. It should not be too difficult as this can be built on top of existing standards for jpeg etc.
Similar to 1994 being “the year of the web”, I think that 2008 is the year of consumer stereoscopic 3D. What criteria will make stereoscopic 3D a success for consumers in gaming and cinema, and who or what groups of people do you see as having the most power to make it happen? Why?
As I understand it, at least some special hardware is needed to view 3D.
If the person who views the 3D does not need to wear anything then 3D will require something very sophisticated such as a color holographic screen, and then I think we’re not there yet. If the viewer needs a special set of spectacles, then he/she will probably not want a number of them but only one. In this respect, video is different from audio.
I do have a number of different headsets: some are earplugs, some are on the ear and two are noise-canceling over the ear. However, all of them are completely interchangeable. All my music sources can use all of my headsets.
The same may not be true for 3D eyewear, and then it will be a disaster. When I buy an iPod I get a pair of earphones with it. They may not be good, but I can start. I can go out to select and buy a headset without having to take my iPod along. If my next Mac has 3D movie capability, then it should also come with 3D eyewear. And I should be able to go buy 3D eyewear without having to worry whether it’s going to be compatible.
When I invite some people over to watch the album of our latest trip in 3D, they should be able to come with their own eyewear. The standard has to be there beforehand, in other words.
I have repeatedly read that you and Tim never patented the web and that you could have made riches beyond your dreams had you done so. Do you think a patent and/or licensing arrangements would have stifled the web’s success? Can similar forces hold back the consumer stereoscopic 3D industry? Why or why not?
Yes, a patent would definitely have relegated us to be a small competitor in a large pond full of incompatible systems. Those systems were already there anyway (Compuserve, AOL, Minitel, …).
I always say that what we did was not provide a service or make software, what we did was set a simple and workable standard: http + HTML. Then we had to work hard to keep that free of adulteration by Netscape and Microsoft. Fortunately the smaller companies joined quickly and supported the idea.
Congratulations on your recent retirement – though I’m guessing this “retirement” is just a facade. What are you really busy with these days?
I’m mainly answering requests for interviews and talks, but I try also to do some of the projects that were shelved during the web years. I really do want to have some fun after more than a decade of continuous hard work.
What key lessons and remarks would you like our members, both industry and consumer, to walk away with from this interview?
Briefly looking back over it, I think the main thing is: simple common standards. Your last few questions made me think how I would have to use the products of an industry, rather than making my own eyewear and tinkering with my own stuff. And that means I would really not want to find out that my new stereo glasses don’t work with my laptop. I may be wrong, but for mass consumption that seems to me to be key.
Dr. Robert Cailliau will be making a third appearance to demonstrate his personal stereoscopic 3D photography. Post your questions and comments HERE, and Robert will answer you in his follow-up interview. A final bonus is Dr. Cailliau has joined the MTBS advisory board, and I am confident his unique experience and mindset will help our efforts to drive the consumer stereoscopic 3D industry forward.