IRT shutdown and Mozilla layoff – or the death of corporate research labs
This is the summer of 2020. The year where Covid-19 appeared. My colleague Rodolphe Fouquet sent me a text during my holidays: IRT would shut down by the end of the year. We knew this could happen but I personally thought it was unlikely.
A few days later Mozilla announces some additional layoffs including the full R&D multimedia team (working on Opus and AV1). We knew this could happen but we thought it was unlikely due to Mozilla’s revenue per employee.
IRT has acted as a common R&D lab for German broadcasters. I love that idea that some experiments and interoperability benefit from being mutualized. As we do a lot of R&D too with the GPAC project we came to meet with several IRT members. I’ve been strucked by the kindness of all the persons I’ve met there. Their behavior was closer to researchers or business facilitators than a company. IRT was very efficient while being able to gather applied researchers, engineers, and a low degree of business oriented toward helping the IRT customers i.e. the German broadcasters.
A few days later we also learnt that Mozilla had laid off a third of their staff as of 2020. The same feeling came over me. Here is why this weakens independent research, innovation, and exposes us to a massive disruption.
These events weaken independent research and innovation, and they expose us to a faster disruption that could make a lot of harm to employment in the media industry.
About corporate research labs
Of course R&D comes as at a cost that most corporate companies business units don’t measure. When business and research are not separate anymore things generally deteriorate.
I think corporate research labs close because the expectations on them are unrealistic. People injecting money expect these R&D centers to behave like long-term traditional businesses with some risk on the short term. They minimize the effects of the long-term collaborations for the ecosystem. They also expect labs to sell products or give them access to services and products for free. But these commercial activities are at best a distraction.
The only way to keep your R&D rate high
Some companies like Apple recognised virtually no spending on R&D as R&D is everywhere. That’s also the approach we took at Motion Spell where R&D expenditure are mostly parts of other commercial activities. That’s the only way we found to keep R&D above 30%. Such a high rate with such an organization means that we take limited risks while being able to innovate constantly.
Delegate, open, mutualize
Corporate research labs often make market refinements over publicly funded research labs discoveries. There are only few places where both meet: IRT, Telecom ParisTech (where GPAC is hosted), or even MPEG where video codecs researchers meet the industry. These places are incredibly valuable for the ecosystem.
Corporate research labs should act as public research labs. Findings should be shared publicly, mutual actions should be taken. Being divided has put some manufacturers in a position to charge developments or licensing fees on every market they targeted. Of course this is done at the expense of users (both broadcasters or consumers).
As mention in the document called: The changing structure of American innovation: Some cautionary remarks for economic growth:
The past three decades have been marked by a growing division of labor between universities focusing on research and large corporations focusing on development. Knowledge produced by universities is not often in a form that can be readily digested and turned into new goods and services. Small rms and university technology transfer offices cannot fully substitute for corporate research,
About the media industry
As we wrote a couple of times, our industry has come to maturity. A lot of operations/mergers happened over the last ten years, leading to the emergence of giants.
A possible disruption
Customers sometimes ask us why we express such an interest in how our industry is structured. We deeply believe there is a growing risk to be disrupted.
Our industry has taken positive actions to move toward more software and more Internet. But it has failed to listen and gather the different voices of its members. The reality is that Microsoft, Intel, Apple, Google, Facebook and Amazon are now parts of our events in ever-growing ways. They make some codecs (AV1), make the transmission (CMAF, HLS), build the platforms (Youtube, Facebook Video, Twich), control the OSes on desktop and mobile…
As the founder and CEO of Motion Spell (the company behind GPAC Licensing) I’ve seen our revenue from the broadcast/OTT industry shift from 90% in 2013 to 55% in 2019. Other customers are from the industry (medical, transport, …), security, social networks, gambling and e-sports.
As mentioned a lot of experiments go (blockchain based video, etc.) occurred. IBC and NAB opened their doors to more startups. But these action remain mainly behind the radar because no actor would actually take the risk of such a change (“One cannot be blamed for choosing Microsoft”). The best outcome for such companies is to be bought and digested by a bigger one. This creates the risk of innovation coming from another unexpected angle (e.g. a successful UGC platform that pays creator with a cryptocurrency for micro-payments).
“The best way to predict the future is to invent it.” Alan Kay