Pankaj Topiwala

  1. Are you or have you been an MPEG member?
    I have been an MPEG member now for nearly 20 years (started in May, 2000, Geneva).
  2. What do you think makes MPEG special?
    There is little question MPEG has had a large impact on multimedia storage and transmission.
  3. What do you think is the most important MPEG impact?
    For me, working in video, I see the impact in video storage and communications to be of the largest impact, from VCDs to DVDs, Blu-Ray, digital TV, and now streaming technologies.
  4. Do you think MPEG is a good conduit for research?
    MPEG has been a great conduit for applied research in multimedia and related fields, and the state of the art has been greatly advanced.
  5. Can you comment on your MPEG experience?
    This is a large question. I’ll focus on video. I have found the collaboration with ITU to be the most fruitful in terms of producing technologies of wide impact. No collaboration is easy, but I have found that the marketplace is most receptive to our joint standards.
  6. Are you satisfied with MPEG standards? and
  7. Do you think MPEG standards are the right choice?
    Since we have 500+ participants, each with somewhat different agendas, such a large group is not easy to manage. And we have produced many standards, not all addressing meaningful market needs. In video, I believe, other than MPEG-4 video, we have produced excellent standards, right for the market.
  8. What do you expect from MPEG in the future?
    This is I believe the most important question. We are no longer driven by technology progress alone, but rather by license-ability. We face a challenge from AOM that we must take seriously. And we have divisions within our contributing companies in terms of licensing that seems to have weakened the prospects for HEVC, and therefore also VVC.  Meanwhile, AOM may face legal challenges, and a legal defense fund has been formed. I realize in response, we have also initiated MPEG-5, with an early release date. It is too early to tell if that will prove to be a valuable second front, or a distraction from our VVC work. Historically, if we had done this earlier, I would have said it was a mistake. In the present era, there are no certainties.  Since a large group of patent holders, as in HEVC, seem to have difficulties settling on licensing terms, it does make sense to create conditions which may allow a smaller, perhaps more committed set of patent holders, that may produce workable licensing terms.  How to manage that with our on-going work with VVC, and the delicate relationship with ITU, is a challenge.  This work area was not jointly conceived, it has apparently now become a source of some distrust and irritation rather than collaboration. That aspect will eventually have an effect in the marketplace, as we will send mixed signals.
    In short, I see many challenges we face. The marketplace wants technologies which we know how to create. But our structure (as a standardization group, but not a licensing group) does not allow us to create all of the conditions for successful acceptance in the marketplace. This has created the conditions for other groups to step in. I’m not sure how to fix this. We have tried to create video standards that are royalty-free; but I do not believe we have the right structure to do so. Our response has been to create a structure in which very few companies can participate in the creation of a standard, in the hope that a smaller number of IP holders will lead to quicker and more acceptable licensing terms. While that may work, it is also partly against our spirit of collaborative design.
    Here is one, out-of-the-box idea, which I put forward mainly to expand our thinking. We cannot create leading-edge standards in video that are truly royalty-free, in my opinion. (AV1 is encumbered by many patents, but is being offered in a rather restrictive licensing structure, and there is a legal defense fund.) We need a way to create a standard that is not free, but is assured to have a limited, acceptable fee. If this were possible, then we could have the conditions that a collaborative design. Such a structure may be open to legal challenges of price-fixing, and anti-trust. The reality is that we are actually working to increase competition, and fair value for consumers, and not against their interests. So I am personally of the opinion that it is morally justifiable. However, the legal challenges are also quite serious, and how to get around those, even if working in the public interest, would not be easy.

Leave a Reply