The Real Origin of Cisco Systems (1999)

Here's a complimentary article from Bill Yeager's perspective: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1407787

> I asked Len why he wanted my source code, and he told me that facilities wanted to take over the router/EtherTIP’s development because I couldn’t dedicate myself to full-time support of the system I had invented and developed with help from Mike and Benjy over the past five years. This request seemed reasonable to me, so I gave him the access he requested and thanked him for his willingness to maintain and improve the software. I didn’t know that Len and Sandy Lerner had incorporated Cisco Systems a year earlier or that Len might have had an ulterior motive: to do a rewrite and then copyright the sources as Cisco Systems’ intellectual property.

> I learned about Cisco a year later when I was called into Stanford’s legal department and told to bring a hard copy of my sources. Needless to say, I was a little nervous. Upon arrival, I was greeted by Stanford lawyer Iris Brest, who explained Cisco’s existence and Len, Sandy, and Kirk’s involvement. She then asked me to compare the Sumex-AIM sources with the EE sources that Kirk had written and tell her if I thought the work was derivative. Most of the EE sources could best have been described as plagiarized or paraphrased: variable names were changed, subroutines were renamed, and large data structures were broken into smaller ones, but identical parts abounded throughout the code. Kirk had added new features and removed others, but the “derivation” was obvious even to Iris who, from what I could tell, didn’t have a technical background. She thanked me, kept my sources, and sent me on my way.

> Just to be clear, I didn’t object to the formation of Cisco Systems or its use of the code I had invented — in fact, I was pleased that work of which I was extremely proud could be used in this manner. However, I did object to the theft of intellectual property implicit in Cisco’s copyright on the sources.

In case people are suspicious/wondering about this story, it is credible to me. I worked with Bill Yundt and he told the story back in 1996. I've also seen the absolute lowest layers of Cisco IOS for 68000's and it certainly appears to come from that era of computing. One especially surprising and interesting thing to me is that it uses cooperative multitasking, not preemptive. This is how systems were written in those days, based on the limitations of early microprocessors. (At the same time in the industry, protected mode multiprocessing existed. But it was in big iron, controlled by IBM, Cray, Unisys and CDC. And those are all of the has-beens now: because technologies like microprocessors, even with their limitations, took over the industry.)

I worked at Cisco 1999-2001, it was my first job out of school. I worked in a group that did network management software, so we weren't touching iOS.

But it was kind of wild at that point there were still company mailing lists where these old heads would argue about iOS internals and flame each other in front of the whole company.

We still had a non-web bug tracking system while I was there. It was an interesting era! The product I worked on did have a web interface as essentially its only UI. We used Java, at some point we used MS Visual J++, and this was before JSPs existed. We used some proprietary templating engine to generate HTML.

Similar to this, RetroBytes (the Youtube channel) did a video on the origin of Cisco recently which is worth a listen at 2x https://youtu.be/NXTdwzjiW7E?si=bCVpmEkyf1UUCfyR

Apparently, the relationships between the executive leadership at Cisco (especially Len and Sandy) were ... volatile. I took a class from one of the top execs in the early days and he had so many colorful stories about his time there.

At Stanford in a non-academic IT department, not long after this was published, we had to beg for departmental firewalls to protect laptops, desktops, and "servers" (but not so much for proper servers). Every machine was essentially on the upper-cased then Internet with a globally-routable IPv4 with almost no Layer 2 IP ACLs. That didn't work so well with negligently-vulnerable Microsoft OS and software. The plus side was that Cisco gear was the standard networking stuff and relatively low cost for hardware and feature packs. Curiously, the housing draw (lottery) was run on a BSD box. About half of the department ran on top of FileMaker Pro (FMP), which ran on Xserves. It was the Windows 2000 and XP/2003 Server machines that were the biggest pains.

Yet another example of how government research drives modern innovation, and how the latest assault on it by the Trump administration will wipe out decades of innovation in the US