At OFC 2019, this writer happened to be sitting next to an analyst, who is with what is officially considered, by certain people in the market, as an influential research firm. I told this individual that at a couple of relevant presentations from Microsoft, including one from Jeff Cox, Partner Director Network Architecture, that I did not hear the term, COBO mentioned – and that our company position was that Brad Booth, Principal Network Hardware Manager, at the cloud player, was totally responsible for pushing that concept. The response was, “Well, you know that Cox is Booth’s boss.” (Of course, I nodded affirmatively, while thinking to myself, “Wow, I am supposed to assume that I was taken to school!”) The reason why I did not ask Cox at that other session about COBO was because I was more alarmed by his insistence that there would not be a need for MultiMode Fiber (MMF)/VCSELs, with his plan to flatten the switch architecture in the data center (even the “prominent analyst”, at least privately, agreed with me that on the very short lengths, MMF would be advisable). Yet, Cox could not resist including “Co-Packaged Optics (CPO) in his answer, which I have generously referred to as “ethereal.” So, altogether, there is a single human being in the space, with evidently the implicit permission of his superior, to keep pushing the COBO model, while everyone else apparently knows that Microsoft, as a firm, has at a minimum, informally and essentially, moved away from that scheme. Nonetheless, the PR effort to keep COBO alive is still going strong, and will likely remain so, for at least a while, taking the optics space into uncharted territory, and perhaps it should now be called something new, like “vague vaporware.”
fibeReality certainly predicted its demise a year ago, but was rightly criticized by fine technologists for being overly expansive and conciliatory. In fact, it could be asserted that engineers held back in stating that we failed to hit the mark on the mindlessness and pomposity associated with COBO, and the utter disregard of a large company putting a single person’s feelings about his deeply-flawed, pet project, over the well-being of the industry.
Still, it must be a humbling experience for Booth that so many people refuse to give up on spreading his dream, despite overwhelming signs to the contrary. We salute him in sticking to his principles, regardless of whether it violates physics principles.
We are relieved that we do not have to wait with bated breath anymore on who will represent Microsoft at ECOC 2019 on the topic of COBO (late update: we noticed that the topic was changed to “Moving Optics Inside,” which hopefully will be somewhat of a good sign). It was only fair that Booth was given that special honor. After all, the combination of Booth’s talk along with those from Lightwave Logic, Acacia Communications, and Infinera may be remembered behind closed doors, as the “Cult Lovers’ Market Focus.” (Perhaps I will be called the ugly American, because I could be accused in favoring OFC, in just calling the US show, a “hyperbole-fest.”)
Naturally, and in all seriousness, with fibeReality being a corporate member of OSA, we felt an obligation to quietly tell the organization, albeit indirectly, that it would be wise to exclude the COBO travesty from next year’s show. However, Booth will be able to take heart in that, unfortunately, we fully expect there will be panels discussing the MSA at OFC 2020. The credibility of the business simply becomes tainted when suppliers are in such a predicament of having to decide whether to devote resources to a mythical standard, in which the only objective of a large customer is to mess around with them, despite them managing to stick around, with the user’s participation in the destruction of an entire ecosystem. (Lastly, this writer is all about the freedom of individuals to empower themselves as long as it does not violate the same rights of other people — it certainly should not include unjustified personal enablement of an otherwise independent, self-sufficient person by either tax-payer funded governments or by shareholder-owned corporations.)
For our fast-growing, totally separate, quick update blog, which is exclusively on fibeReality’s LinkedIn page, please follow us here.
As always, fibeReality does not recommend any securities, and this writer does not invest in any companies being analyzed by us.
[written by Mark Lutkowitz]