Kip and Chris’s time at the California Institute of Technology is what these two astrophysicists have in common. They both worked at CalTech’s physics and astronomy department during the 1990s. Chris recalls, “[Kip Thorne] was very kind and wrote most of the one paper that has my name on it. I had done some initial computational astrophysics studying the structure of supergiant stars that typically have degenerate matter cores, but the hypothesis was that they could support a stable neutron core as well.”
What Kip and Chris don’t have in common, is working at the upstart startup, Cerent. However, talking about stable cores, Cerent’s management system would ultimately feature a stable core, rooted in CORBA (Common Object Request Broker Architecture). Chris was instrumental in this program in support of the Cerent 454 offering, as well as its progeny.
Fiberlane was formed in 1997 with Vinod Khosla and Raj Singh leading the charge. Concurrently, Cassini was launched safely by NASA toward Saturn on October 15th of that year, a mission that is coming to a close as we write these words. Like the ‘454,’ the Cassini robotic probe has had a long, productive life.
Gamma Rays and Startup Competitors
“Any photon is entitled to call itself a gamma-ray,” wrote astrophysicist Virginia Trimble in December 1998 , “but what is meant here are ones from PeV or TeV down to 100 keV or thereabouts (that is, photons captured by devices funded by gamma-ray programs rather than X-ray programs).” Similarly, any optical transport startup felt entitled to call itself an innovative platform to replace or augment the existing SONET infrastructure in America, but only Cerent reached astronomical heights.
Trimble humorously discussed Gamma-Ray Bursts (GRBs) in her year-end summary paper, noting, “GRBs have been all over the place in the literature (78 papers entered in our notebooks in the index year, excluding IAU Circulars) and in space, with associated redshifts ranging from 0.0084 for the event 980425 to 3.42 for GRB 971214.” 
In the parallel universe of optical transport, Cerent’s early success inspired copycats that were literally popping out of the woodwork. But the billion-dollar market was only so big. Copycat products promoted by imitating companies began to appear all over the place in the literature too, throughout 1998, 1999, and 2000. These wannabes included the likes of Redback (using the Siara Systems engineering team from the other half of the Fiberlane splinter), Metro-Optix, and White Rock, companies who came and went; to the data-centric boxes like those from Alidian and Chromatis; to the wavelength-centric boxes like those from Kestrel, ONI and Sycamore, all of which, except for Sycamore, never found sustained traction.
However, Cerent’s MSPP value proposition was rooted in a SONET/SDH-based platform that would fit into existing metropolitan optical networks (at any bit rate and in any network configuration) while supporting new revenue-generating services. Its initial acceptance was driven by slashing OC-48 configuration prices in half and delivering on the promise of footprint savings, reduced power consumption, and plug-and-play service flexibility. Being first to market was one reason for success, but it doesn’t always guarantee success. In this case it did .
The Redshift History of Star Formation And Startups
“Once upon a time, there were no stars,” wrote Trimble in 1998. “Now there are a great many, mostly not very young. Necessarily, then, the star formation rate has been fairly large at some point in the past.”
The optical transport backdrop of the late 1990s was growing rapidly with the advent of Nortel’s OC-192 product offering introduced coincidentally with the proliferation of long-distance companies (remember WilTel?) and the appearance of the backbone-breaking demand of the expanding Internet traffic. The “not very young” established players of the market included Nortel, AT&T Networks (later Lucent), Alcatel, Fujitsu, and others of that telco-infused ilk.
“[While] theorists tend to think of star formation as a function of time, since they calculate the progress of mergers, bursts, and other dynamical, star-driving processes in years,” according to Trimble, “[Astronomical] observers, on the other hand, tend to report things in redshift units, since that is what is directly measured.”
Similarly, industry analysts covering all things optical thought of new startups as a function of time too, but not it in terms of years, rather in months. The process for a startup to secure funding in 1998 seemed to be, simplistically, to come up with an idea (theorist-like), develop the concept in powerpoint, and present it to an eager venture capitalist. Very quickly the funds were granted in an era of irrational exuberance. The fortunate, funded startups  plotted their exit strategy as observers of the industry “calculated the progress of mergers, bursts, and other dynamical processes.” Whereas “things” used to be reported in terms of revenue units, the late 1990s saw “things” reported in valuation units.
Protection from Comets and All Optical Comers
“Jupiter, [our Solar System’s gas giant], may be guarding us from frequent bombardment by comets,” wrote Trimble, clearly with the 1994 Shoemaker-Levy 9 strikes of the Jovian surface in mind. With tongue-in-cheek, she adds, “We had no takers on last years offer of comet insurance, and are thinking of tailoring the new prospectus to a Jovian audience.”
Cisco, the industry’s Internet giant, protected Cerent from the dual bombardment of an impending dot.com bust and the telecom meltdown by having acquired it in 1999. Cisco’s prospectus would be augmented to tout Cisco as the disruptive innovator in the optical transport space and use Cerent’s team as an insurance policy against failure in the service provider market segment.
Naturally, the company’s investors and Cisco-watchers pressed Chambers for details as to how Cisco would reach this goal. He added, “You'll want to watch our momentum with the 15454 and then if we come up with a lower end of that product, which we probably will, [watch] how we do within that lower-end product area.”
Then, as Chambers had hinted, on January 31, 2001, Cisco Systems officially launched the ONS 15327. Like the Steve Jobs-run Apple, with products available as they are launched, Cisco’s “mini-me” product was ready to ship to Cisco’s carrier customers that day.
But With All of this Excitement, Why Did Chris Eich Leave Cisco?
With his astrophysics background coming to the fore, Petaluma-based Chris explains why he left Cisco: “I could see the accretion disk forming around San Jose and more and more of the technical decisions were being made out of San Jose and the leadership was being changed out.”
With a subdued injection of sarcasm and a constrained smile, Chris adds, “I lost track of all the management people that cycled through after Gary [Baldwin] left. You could certainly see the writing on the wall [for Cisco’s Petaluma operation].”
Chris concludes, “I kind of caught the startup bug when I joined Cerent as employee number seventy-five or so, and so I joined Enphase Energy as employee number nine. ”
And the Universe continues its motion, just as startups come and go and their entrepreneurial-minded employees come and go.
 Astrophysics in 1998, by Virginia Trimble and Markus Aschwanden, ASTRONOMICAL SOCIETY OF THE PACIFIC, 111: 385 – 437, April 1999.
 “A VERY crude summary is that, up to about 1998 October 1, 18 X-ray afterglows had been caught” Trimble added. “Half of these had yielded optical identifications. When the optical transient faded, there was always some sort of host galaxy around, with the GRB having been fairly near the center and the hosts not giant Es or Ss, but rather feeble things, like the galaxies littered over the Hubble Deep Field images. Two or three also had rapidly varying radio counterparts, with much of the early variability being due to interstellar scintillation, which decreased when the sources expanded to about 1017 cm in size. And, finally, it is nigh on to impossible to extract this minimal data set from anything in the archival literature, and we have cheated rampantly, making use of an excellent review talk given by Neil Gehrels at a conference early in October . . . Truth (whether godly or satanic in nature) is widely held to be in the details. This subject is, however, very far from being ready for review at either a detailed or divine level. It is safe to say that the bursts are not all precisely the same sort of thing.”
 Another success factor is the ability of startups to navigate economic downturns. While companies like Cerent and Siara were part of Cisco and Redback, respectively, other startups that went public after 2000 were not so well insulated from the slowdown in capital spending or the muted challenge to established service providers from competitive carriers. These troubling trends began to appear in 2001, reducing the revenue projections of many optical transport startups. Their built-in cost structure and continued high cash burn rates could not tolerate the much longer road to profitability. Any service provider capital spent was more likely to find its way to established vendors instead of start-ups, especially those new companies with untested products that had yet to be evaluated.
 One of those startups that did quite well during the 2002 and 2003 downturn was Turin Networks, a company also funded, in part, by Telecom Valley’s Don Green and initially headed by one of his protégé’s and founder John Webley. Turin would continue to build momentum in the MSPP space at Cerent’s, now Cisco’s, expense. After all, competition is healthy for the marketplace.
 Enphase is based in Petaluma, California. I interviewed Chris Eich at the Enphase Energy campus on May 28, 2013.