All the acronyms a man could invent
|Leonardo Chiariglione, winner of the IBC John Tucker Award: "My only job title is head of the multimedia services and technology division at CSELT. For the rest I consider myself a manager of worldwide technology convergence."||Leonardo Chiariglione: "Why on earth should there be a one-to-one match between the resolution of the camera and of the display? This concept is simply a legacy of analogue TV and a legacy of the 'occupy the display real estate' philosophy."|
This year's international honour for personal excellence in the service of the broadcast industry, the IBC John Tucker Award, has been won by Dr Leonardo Chiariglione, head of multimedia services and technologies at the Telecom Italia R&D subsidiary CSELT based in Torino, and convener of the MPEG Committee. He will receive his prize of £5,000, an official scroll and the trophy from IBC co-founder John Tucker on stage during the Sunday Spectacular.
The citation Tucker has chosen for Chiariglione's scroll will read, 'In recognition of his outstanding leadership and unstinting personal contribution to the audio-visual media industry.' Adding his personal tribute, Tucker said, "Leonardo is a man of enormous drive, much respected across the whole media industry for his initiative and leadership. He gets things done, and on the way makes some enemies. He is a man of our time".
Chiariglione was born in Almese. His education up to MS level was in Italy, where he graduated in electronic engineering from the Polytechnic of Turin. He won his Ph. D. degree from the University of Tokyo in 1973. He joined CSELT (Centro Studi E Laboratori Telecomunicazioni) in 1971, and recorded the first of many carcer milestones four years later with the development of a RAM-based video simulator. His most recent achievement has been his appointment as esecutive director of the Secure Music Initiative (SDMI).
In conversation with Daily News editor George Jarrett, Chiariglione revealed his role as the enabling force behind many of the industry's most crucial standards initiatives over the past 20 years, not least the success of MPEG in its various numerical identities. The origins of his unique personal drive to serve the media world's perpetual need for agrecd stantìards can be traced back to his school days.
"I believe my motivation originated from one of the visits that I used to make in my high school days to the USIS office in Turin," he said, "They had a good library, and my first reading of E L Masters' Spoon RiverAnthology happened there.
"That day I saw a poster of President Kennedy addressing 'the citizens of the world': it was a new concept for me and it is still one of the drivers of my work."
Asked what career achievement has given him the greatest satisfaction so far, he selected the first and last of his initiatives, but not because all the others have yielded no saltisfaction! "Being able to convene 50 people from everywhere in the world in the unknown-to-most city of L'Aquila in the mountains of central Italy for the first workshop ori HDTV (1986) and being able to create the conditions for people to introduce the first specifications for secure digital music in four months are great satisfactions indeed," he said.
Those that do not know CSELT might wonder what it does, and how come Chiariglione has been accorded so much time and scope to work on a personal agenda that is far wider than the typical commercial desires of any norrnal company. "So far I have been operating in a world where companies were farsighted enough to devote part of their profits to invest in the future," he said.
"My company, CSELT, has allowed me to kick off and stay at the helms of a number of technology developments that have now impacted its business. But if you think of research institutions like the old Bell Labs of AT&T, the T J Watson Research Center of IBM, the Nat Lab of Philips, and the inventions they have bestowed on mankind, I would say that my actions have moved very much in the wake of that tradition.
"Unfortunately, the external conditions are changing and companies are becoming more stingy - or less far-sighted. The world is changing, but I am not sure it is for the better. Two of the technologies that are shaping our world, signal processing and the Internet, are the result of very far-sighted investments by the Bell Labs and the US Government. At times I have the impression that we are squandering the assets that our forefathers have handed over to us and that we will leave nothing replacing it to the new generations."
When people climb onto stages to collect awards, many of them thank loved ones and mentors. The Dr's good cures for CCIR, IVICO, DAVIC, EUREKA and MPEG have not left him with a great list of professional heroes. "What I can do is look back and thank my parents who instilled in me the thirst to know, and my professors at the Salesian high school of Valsalice in Turin who inculcated in me the broad minded attitude on which the message of President Kennedy could take root. In particular I remember Professor Bonello who helped me adapt from the Classic Lyceum to the Polytechnic of Turin, Professor Boella who helped me get my first stage at KDD labs in Japan, Professor Miyakawa, my supervisor at the University of Tokyo, and lastly Mr Mossotto, the director general of CSELT."
A feature of Chiariglione's professional life is all the acronyms a man could possibly want. The two that seem to have the strongest meaning to him are MPEG and HDTV, so we moved on to discuss his 1986 HDTV hide away and eventually arrive at MPEG. First, why did he want to stage that workshop?
"The idea came to me after the failure of the then CCIR to agree on a common standard for HDTV, and that was my first experience in setting up an international cross-industry concern," he said. I thought that because the "political" approach in defining HDTV had stalled we needed another approach. That was putting together technical people from all the companies with a stake in this new medium and have them discuss the technology needed to make HDTV real, independently of the shape that it would ultimately take.
"That is why I never worked for a specific format - that many lines, that many pixels, that many frames per second. Actually, I am of the opinion that digital is in direct contradiction with such a concept of format. For a decoder what matters is the number of computations per second and memory size: this concept I implemented in the MPEG standards."
Before we got to the development of MPEG, the Dr had thoughts on Europe's shunning of HD, and the possibility of a change of heart. "Broadcasters were the first to discover that the first task is to occupy the entire screen of the consumer. For IT people it took 50 years to get to the same conclusions and now the 'first screen' war rages.
"I see a future for HDTV not so much in the form of content obeying a particular format, but as a display device of sufficient resolution, say 2 Mpixels and hopefully progressive. That display will be the real estate, owned by the consumer, on which he can autonomously decide to compose and present pictures with different resolutions from different sources.
"Why on earth should there be a one-to-one match between the resolution of the camera and of the display? This concept is simply a legacy of analogue TV and a legacy of 'occupy the display real estate' philosophy."
Having never been a great fan of compression, I wanted to know how far Chiariglione believes MPEG can take the industry, and how long its natural life expectancy might be. I was expecting a big big fan. "MPEG-2 does wonderfully the job for which it was designed, i.e. compressing progressive or interlaced pictures. Almost five years after its formal approval as a standard, there is nothing even remotely in sight that produces anything 'better' that could justify considering a change," he said.
"What MPEG-2 does not do, obviously, is what it was not designed for. How do you compose different pictures? How do you synchronise them? How do you add hot spots? These and other features would be interesting for a gradual introduction of interactivity, possibly local, in a broadcast environment.
"By this I do not mean mimicking the many text-based extensions of the Web. This has been attempted for years but has apparently led nowhere. What I mean is interactivity as a natural extension of the television paradigm. For this the technology exists; it has been developed by the same group that developed MPEG-2 and it is called MPEG-4.
"The two technologies can go on coexisting for a long time, each playing the role for which it was designed."
Chiariglione recommended Nick Wells to head the Pro-MPEG Forum when he was approached with the idea, no doubt one of the deeds that led one industry voice to dub him the 'Godfather of MPEG' when assessing his parental role among the MPEG missionaries.
"I looked in Webster's and found the following definitions - 'a man who sponsors a child at baptism ... and may thereafter take an interest in the child's upbringing and welfare'. I think this is a fair description, but I would exclude the second definition, - 'a man who assists at the Jewish rite of circumcision'. "
Given his role in establishing MPEG, Chiariglione's views on the rival DVCPRO compression technology, and both group's show floor claims for 'inter operability' advantages were worth waiting for.
"The DVC specifications are technically sound and the products using them are good, as much as MPEG-based products are good. But this is not the issue," he said.
"It is a pity we have two technologies MPEG and DVC - that are virtually the same: both are based on DCT, but incompatible. Because I knew what the future would prepare for the industry, in 1994 I wrote a letter to the DVC chairman suggesting that we consider an alignment of the two technologies. The response that I got was that DVC was designed for consumer devices - exactly as MPEG-2 by the way.
"Now DVCPRO presents itself as a technology for professional applications much as MPEG-2 is also used for professional applications. The losers of this battle that should and could have been avoided are the users. And this loss on users' part does not serve the ultimate needs of the companies that back one or the other camp. It's another Beta vs VHS war.
"There must be a winner," he added. "Imagine a cameraman using MPEG, a studio using DVCPRO and the end user using MPEG again. It is the recurrence of the old adage: to make the system work you need the interface of the interface of the interface. 'Oh, will they ever learn,' sang Joan Baez in another context."
A scan of the Chiariglione CV brings in so many bodies, like FIPA, OPIMA, DAVIC, CCIR and RACE, as well as MPEG. Like any good dance through the acronyms, I wanted to know how one thing, each initiative, led to another.
"In 1988 I started MPEG because I realised that after decades of research in audio and video coding the technologies were ready to be converted to practical applications. Standards were needed to achieve this goal, but where should those standards be developed?
"The question was relevant because one of the curses of standards bodies is that they are organised along industry lines - broadcasters in CCIR, telcos in CCITT, CE companies in IEC, IT companies in JTC 1, the rest of the industries scattered in other parts of ISO. But digital audio and video is - and the IBC attendance is proof of what I say - a basic technology for everybody. My choice fell on JTC 1 because at that time the attention of the IT world to digital audio and video was marginal and the standard could be developed without too much interference.
"At the beginning of 1994 I established DAVIC because I realised that the MPEG technology would not have been sufficient to provide a complete solution to my dream of end-to-end interoperability across countries or regions and application domains. OPIMA (Open Platform Initiative for Multimedia Access), established at the beginning of 1998, is an initiative with the goal of defining security in the platform in such a way that anybody, provided he has the appropriate credentials - and here I do not mean a proprietary set-top box - can get access to content. This is a fundamental technology that gives content providers the ability to offer their content to anybody in a protected form.
"FIPA (Foundation for Intelligent Physical Agents) established at the beginning of 1996, aims at defining the language and the environment in which intelligent pieces of computer code, representing different entities - agents - can talk to one another to, say, negotiate a deal or to accomplish a complex task together. This is another key technology for a future where billions of users will try to access vast content offers through a variety of delivery mechanisms with associated performance levels and costs."
Broadcasters are not playing fair with viewers over the number of compression artifacts that jump out of the TV screen, and it is my expectation that the viewer is going to have to develop a blind trust in Tektronix picture test products if DTV is to impress consumers. Having heard Michael Isnardi's wonderful MPEG lecture, I had to ask the Dr what kind of devil he let lose, and whether he keeps the key to its cage.
"You are seeing the effects but you do not inquire into the causes. If you go to the fruit market and buy apples that are described as good but you find later they taste bad, next time you will not buy apples from that guy. Strangely this simple precept finds no home in the digital TV market.
"The DTV market is just a collection of flefdoms, and consumers in one of them have a hard time if they want to escape. This is a mockery of competition, with no regulators, no consumer organisations caring about such a distortion.
"Because MPEG is so flexible - this I suspect is the reason for you using the word 'devil' - it can be used to deliver component studio quality at 9 Mbit/s, composite studio quality at 6 Mbit/s, DVD quality at 3 Mbit/s (VBR) and VHS quality at 1.5 Mbit/s. Because there is no competition, it is only obvious that providers find it more convenient to pack as many programmes as they can, and if the resulting quality sufffers, too bad!
"If consumers had access to set-top boxes where security is not at the service level but at the platform level - I mean the OPIMA solution here - you would instantly see holders of franchises become sensitive to consumers' needs and the problem you mention would automatically disappear. Users would have the ability to make their own trade offs, scraps at a discount, good pictures at the right price, instead of being forced to swallow what is given them."
A lot of IBC visitors will have heard of SDMI, and will want to know more about specifications 1 and 2. Chiariglione has no lesser enthusiasm about this initiative than about the birth of MPEG. "SDMI is developing a very important piece of technology to solve a key problem, one that the broadcasting world has started to struggle with since the advent of digital," he said. "And that is: how do you prevent content in digital form being copied millions of times, thereby reducing its value to zero?
"This has become crucial with digital audio because even with the low bit rates of the web today, it is possible to exchange high quality audio files. On the one hand this becomes a potential source of worrisome large-scale piracy but on the other hand it opens virtually unlimited ways to sell content to users.
"SDMI is about providing solutions that solve the problem of protection while retaining the flexibility of implementation of new business models made possible by electronic commerce. The future of media will be shaped around these technologies. I am in the process of writing a paper that should shed some light on the future of media as I see it. Find it at http://www.cselt.it/leonardo/paper/wipo99/"
Having had little success with the concept of 'mentors', I wondered if Chiariglione had trusted in the friendship of allies during his technology frontier work.
"I do not like the word 'allies'," he said. "For me this carries the meaning of people grouping to fight other people. I never did that.
"In MPEG- 1 times I refused to consider the work we were doing as a killer of DVI and CD-i. I talked to those companies a convinced them that MPEG, far from killing their ideas of a product, was simply a help in solving a problem - by eliminating the unknown of the compression technology. Companies had better concentrate on the really important part - the platform and the applications - and forget about a piece of silicon that could be purchased from an industry specialising in those chips.
"In the hot MPEG-2 times I refuted the slander made by some that the work were doing was the killer of D2-MAC and MUSE. We were simply providing technology for people in need of it and, if the number of equipments exhibited at IBC is any proof, there was a need for it! But the standard not only solved known problems but it gave rise to new products - like Video CD and DVD that have taken the CE world by storm.
"If you want to use the word 'allies as 'partners', then I do have thousand enthusiasts who have believed in my message and have produced these wonderful standards that are changing our life. I have also benefited from the help of tens of good and dedicated chairmen of my committees who have organised the technical work. These have been my 'allies'."
In an earlier conversation, Chiariglione had said he did not "automatically' go to shows like IBC. He has been to both IBC and Montreux in the past, but even his level of commitinents has to be linked to his at CSELT.
"I consider broadcasting as an integral part of multimedia. Broadcasting, the point-to-multipoint delivery of information, is successful today because the state of development of interactive networks makes it competitive. With the future abundance of offers, I see that the specifics of delivery will become completely opaque to the user. The user, or better his agent, will simply negotiate the terms of delivery - now or delayed, streaming or file transfer, high bit rate or low bit rate, interactive or not, guaranteed delivery or best effort."
One school of thought behind IBC believes that producers and creatives are now driving the market and making the buying decisions, and that engineers and technologists have become less crucial. Another school believes technologists still hold sway and set the agenda. Meanwhile the industry is about to handle almost total control to the consumer and adopt the Internet as its new 'transmission' medium. Looking at the near horizon, which camp does Chiariglione see holding sway - the technologist, producer or consumer?
"I think both views are correct in the right framework. On the one hand one could say that the only players that matter are those producing content and those consuming it. All the rest is made up of 'disposable' intermediaries - including my employer. On the other hand, technologies keep on providing more and more ways to make content richer - take the combination of MPEG-2 and MPEG-4 and see MPEG-7 and SDMI, offering content in new innovative ways.
"I believe a mutual understanding and respect of the different roles is fundamental to making the industry progress."
Along with the cinematographers Carlo Di Palma and Vittorio Storaro, Chiariglione is fronting an impressive Italian presence in Amsterdam. I wondered if he would provide a personal viewpoint of where his country sits within European and world-wide efforts to keep standards the driving force of technology acceptance, and in the adoption of all the proposed consumer delivery vehicles. "My registered address and workplace are in Italy but the scope of my work is world-wide and, as I said before, I consider myself as a citizen of the world. I also consider myself the standard bearer of a different way of making standards that retains the good side of the old ways while avoiding their many evils: see http://www.cselt.it/leonardo/paper/standardisation.html for a paper describing my ideas on the subject. Basically my recipe is that standards battles should be fought in standards committees managed like companies, not in the marketplace with consumers' wallets. My great fortune is the personal support of the director general of CSELT in my endeavours.
"Researchers of CSELT work hard to provide our parent company with new ways of staying competitive. My collaborators in particular have some new products that should soon bring a new attractive face to many broadcast and web-based services."
The only things left to ask Dr Chiariglione were whether there is a standards effort he now regards as a waste of time, and if he has one more initiative left in his ambitions. This mere mortal would want to spend more time at the vineyard he inherited from his grandfather.
"No, I mean not those I have been involved in. Even those initiatives with less success have always brought something good to me and, I hope, to those around me in those endeavours. If you read the paper I quoted (above), however, you will see which standards I consider not just a waste of time but killers of those industries that first sponsored and then mismanaged them.
"Something may happen soon. Watch my home page: http://www.cselt.it/leonardo/.