Sunday, February 26, 2012

Bell Labs: Another View

The NY Times had a piece praising the way Bell Labs innovated. I beg to differ. First I was at Bell Labs from 1964 thru 1972, at various times and at multiple locations. I was there from undergrad days until just after my PhD from MIT. From 1986 through 1989 I was also Head of R&D for what was then NYNEX and is now Verizon. Thus I speak from firsthand experience, more than anything the author of the piece can do.

The author states:

Why study Bell Labs? It offers a number of lessons about how our country’s technology companies — and our country’s longstanding innovative edge — actually came about. Yet Bell Labs also presents a more encompassing and ambitious approach to innovation than what prevails today. Its staff worked on the incremental improvements necessary for a complex national communications network while simultaneously thinking far ahead, toward the most revolutionary inventions imaginable. 

I could not disagree more. In my opinion based upon a decade of direct presence and exposure I contend that Bell Labs is the antithesis of how research and technology development in a corporate world should be done! In fact if one follows that model one will fall into a world which we no longer live in. It is a world devoid of creative destruction, a world devoid of any truly competitive innovation, a world where we would have a very small fraction of what we have today.

The author also states:

He personally helped design a building in Murray Hill, N.J., opened in 1941, where everyone would interact with one another. Some of the hallways in the building were designed to be so long that to look down their length was to see the end disappear at a vanishing point. Traveling the hall’s length without encountering a number of acquaintances, problems, diversions and ideas was almost impossible. A physicist on his way to lunch in the cafeteria was like a magnet rolling past iron filings.

Now I spent time at Murray Hill, Whippany, Holmdel, and Indian Hill, I also was at the West Street office on my first days. In reality it was all too often a 9-5 location with Chess Clubs, model airplane clubs, amateur radio clubs, bridge clubs, which occupied many hours in excess of lunch. In my ten years of exposure I failed to see much of what the author contends. My conclusion is a generality, but one based upon a broad exposure.

Now to the facts. Bell Labs was set up to support Western Electric, the manufacturing arm of AT&T. Western was the sole supplier to the Bell Operating companies, BOCs. ATT was a Government sponsored monopoly exempt under the law from antitrust restraints along with baseball. Bell Labs was thus a way to develop technology for the operating companies and also to create patent rights to prevent any other entrant into the business. It was a Government sanctioned monopoly which effectively insured telecommunications related technology development was suppressed everywhere except in Bell. Somehow the author seems to miss that point.

Now a second fact, the BOCs and ATT had a different profit making rule. Unlike the normal market where the price is set in an open free market place by supply and demand, and profit was revenue less expenses, the Bell equation was materially different. They, namely the BOCs and ATT, set rates, actually the Government claimed to do so but in reality, in my opinion, ATT told the Government what to do, and there were times of some disagreement, but eventually in my opinion ATT got what they wanted.

The rates and economics of the business from my perspective generally worked as follows:

  1. Revenue was set based upon rates.
  2. Revenue equaled costs, operating and depreciation, plus a return on the capital plant investment.
  3. Profit then was the rate of return on capital, usually somewhere between 12-18%.
  4. Costs had nothing to do with profit, and profit was maximized by designing the least efficient means of production, namely the more capital per customer the more profit.
  5. There was no incentive to reduce costs or improve technology.
  6. It was a monopoly.

Thus Bell Labs was motivated to get as big as possible and to be as inefficient as possible. However the PR issues were at the fore, so to keep the Government at arms-length they publicized what they did in research and did a great deal of Government development work. For example they did work on undersea sub detection and the Nike anti-missile program.

The author of the article praises all of the Labs development. Let me make a few corrections:

1. Internet: According to Bob Kahn, as I recall having discussed with him, when he was head of IPTO at ARPA he went to Bell Labs to seek their help to deploy packet switching by utilizing some of their modem designs and networks. Bell Labs management, there were a great number at the meeting, which was all too common, informed him that they would take an exclusive contract from ARPA and design and build what Bell thought was right and that Kahn could watch the results. Fortunately Kahn rejected the exclusive deal and in his brilliant way created the core group who “invented” the Internet, despite Bell Labs! I was fortunate enough to play a small part in that effort when I was at Comsat, getting the first satellite connections operating.

2. Satellites: John Pierce boldly published a design for dozens of satellites as necessary to perform as a communications net of limited capacity. He stated that it was technically impossible to have a synchronous satellite. Believing Pierce Comsat was funded assuming dozens of satellites and launches. Hughes soon thereafter launched the first synchronous orbit satellite and only 3 were need for all the world! Pierce was proven wrong as was Bell Labs. Perhaps the Pierce design was consistent with the massive capex overspending as was pandemic at Bell. Harold Rosen at Hughes was the driving force of this new world and he rather than Pierce should have the recognition. Again the author seems to have missed this point. Bell launched Telstar, and then just withdrew as competition arose.

3. Digital Switches: Bell Labs refused to move to digital switches, they had developed the No 1 ESS, a project I had worked on, and wanted to allow a "normal progression" as I recall. They thus stalled. The Chairman of ATT at the time, as I was told, was frustrated and he went to Bell Northern Research, the Bell Canada arm, which AT&T at the time owned, and asked them to build a digital switch. That switch became the basis for Northern Telecom, one of the most advanced switches for a few decades.

4. Modems: In the early 1980s, with the advent of PCs, companies such as Telenet and Tymnet grew and modems were need. Bell refused to do this because it would reduce costs. Instead a small company called Hayes built one of the first digital modems to work on these separate networks. It allowed what became the Internet to grow.

5. IP Networks: The IP based networks came from small companies such as Cisco, and wireless came first from Motorola and then new entrants such as Qualcomm. Telecom as we know it today grew despite of Bell Labs not due to it.

The author, in my opinion, is totally blind to where true progress was made, it was made with the entrepreneur, not the massive corporate research center. I would argue that Bell Labs was a major drag on inventive elements in telecom. It was Codex and modems, Cisco and routers, and many other entrepreneurial companies which lead the way. Entrepreneurial companies work in a Darwinian fashion, success is rewarded and failure falls away. Bell Labs, for many generations, in my opinion, and based upon my experience, actually thwarted development. Perhaps that story may someday be told, not the one sided tale contrived, in my opinion, from some PR machine.