[ih] The linux router project and wifi routers

Fri Nov 4 17:20:16 PDT 2022

In the 84-86 timeframe, Codex built a VME 68K router for Project Athena. I know Codex and have heard that others (including BBN) were  advocating building routers. but marketing consistently argued there was no market for one. We could all see they were wrong but had no data to prove it.  As always, marketing had 20-20 vision in the rearview mirror.

> On Nov 3, 2022, at 14:13, Bill Nowicki via Internet-history <internet-history at elists.isoc.org> wrote:
> 
> 
> Yes, the details might not have been public at the time, since it might have looked bad. What I meant was that about once a year the claim by Cisco marketing that they "invented" the multi-protocol router comes up. Usually, Noel Chiappa points out that MIT had one earlier (Stanford even had at least two of those, based on the PDP-11 and MIT-written software) before then. I see such a discussion on this list in 2017 for example.  And yes, most of the early 1980 efforts used concepts from the Xerox PARC PUP architecture. PARC was on Stanford land, a short bike ride up the hill. I was a summer intern at Xerox, and bunches of us went up to the "dealer" seminars.  There were various meetings with people from MIT, CMU, ISI and other DARPA contractors, where ideas were freely interchanged.
> Cisco's first product, by the way, was an Ethernet adapter for the Massbus on a DecSystem-20, but re-using the router hardware and software gave them a head start on the router business. By late 1986 it became a three-way discussion between Sun, Cisco and Stanford office of technology licensing (OTL). It was complicated by the fact that Andy had tried to license the original Stanford design (funded by NSF and DARPA), but OTL said patents would be hard since it was a modular design with standard parts! There was at least one patent on the graphics board, but Cisco did not use the graphics board. Andy Bechtolscheim had been at the Sun Microsystems company for a while (I was too by then). His hardware design had diverged from Len Bosack's, and Cisco ended up totally rewriting their software without any Bill Yeager code. 
> The director of the AI Lab, Lester Ernest, mostly was concerned that Len had been collecting his Stanford salary and using the facilities he supposedly managed to design the Cisco variant. Les has published a bit out this informally, including an email thread from the time. Most of the bad feelings went under the bridge due to the large discount that Cisco and Sun gave to Stanford, making the university a flagship beta customer. Cisco was trying to position themselves as the "high end" with high list prices and discounts negotiated to each customer so they each thought they were getting a great deal.
> If nothing else, there were follow-on "gifts" (which could be tax deductible!) including endowing a professorship. I did find these:
> Cisco endows professorship in information technology
> 
> 
> | 
> | 
> |  | 
> Cisco endows professorship in information technology
> 
> 
> |
> 
> |
> 
> |
> 
> 
> https://www.quora.com/Did-Cisco-actually-invent-the-modern-router-Or-did-they-just-perfected-it
> 
> 
> Bill
>    On Thursday, November 3, 2022 at 09:53:39 AM PDT, Barbara Denny via Internet-history <internet-history at elists.isoc.org> wrote:  
> 
>  At least one of those people worked at SRI before moving to Stanford and then Cisco.  There were other sources of early technology transfer.
> Later on I was also asked by the military to try to get early commercial router vendors to support one protocol they were interested in.
> barbara
>     On Thursday, November 3, 2022 at 08:54:55 AM PDT, Craig Partridge via Internet-history <internet-history at elists.isoc.org> wrote:  
> 
> Perhaps helpful with the archaeology.
> 
> Several of the single digit employees at Cisco were former Stanford IT
> folks who had worked on the routers.  And the routers were actively used
> for research -- people were periodically making enhancements based on new
> research and then getting them deployed in the Stanford routers.
> 
> I remember one person who leveraged the routers for research despairing as
> all the folks who had supported his research were now at Cisco.
> 
> So, and purely speculating, there may have been concerns about a continuing
> transfer of knowledge/ideas post any initial license as Stanford employees
> who were engaged in improving the Stanford devices then went to Cisco, with
> improvements in their heads.
> 
> Craig
> 
> On Thu, Nov 3, 2022 at 9:07 AM Guy Almes via Internet-history <
> internet-history at elists.isoc.org> wrote:
> 
>> Dave,
>>     The word I heard went into some detail on the discounts, but I recall
>> wondering why no mention was made of equity.  I hope your memory is
>> correct and that Stanford did get some equity.
>>     So, in itself, the "Stanford thought about suing them but didn't"
>> story is not very interesting.
>>     The more interesting thing would be why they were upset at cisco.
>> Particularly given Tom's account stating that cisco had received a
>> license during 1986.
>> 
>>     By the way, I still recall my initial visit to cisco in March 1987.
>> At the time, all of cisco was in a suite of offices and lab space on the
>> second floor of a nondescript building in Menlo Park.  Len Bosack met me
>> and stressed (a) the advanced nature of their routers vis a vis the
>> competition and (b) his ambition for cisco.  Even though the office
>> suite was loosely furnished, his Board Room included a map of the world.
>> 
>>     We ended up placing a huge (seven routers) order with cisco and I
>> never regretted it.  It was neat being able to call cisco to report a
>> bug, getting a sngle-digit-employee-number person to answer the call and
>> have a new version of the software without the bug by close of business
>> that day.  Needless to say, that didn't last for more than a year or so.
>> 
>>     But I also recall friends at Stanford being very critical of the
>> cisco-Stanford relationship during this period, so it's definitely
>> Internet history of interest.
>> 
>>         -- Guy
>> 
>> On 11/3/22 9:05 AM, Dave Crocker via Internet-history wrote:
>>> On 11/3/2022 5:52 AM, Guy Almes via Internet-history wrote:
>>>>   During spring 1987, however, when I was shopping for routers for an
>>>> NSFnet regional network, I was made aware of quite a bit of
>>>> uncertainty about whether Stanford was going to sue cisco over some
>>>> Intellectual Property issues.  Evidently there was a settlement that
>>>> gave Stanford a discount on buying cisco routers.
>>> 
>>> Adding to the rumors...
>>> 
>>> At the time, I heard all of the above, with the closure being that the
>>> settlement was Stanford's getting some cisco shares.
>>> 
>>> d/
>>> 
>>> --
>>> Dave Crocker
>>> Brandenburg InternetWorking
>>> bbiw.net
>>> 
>>> --
>>> Internet-history mailing list
>>> Internet-history at elists.isoc.org
>>> 
>> https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/internet-history__;!!KwNVnqRv!BhF0ReQHMaqoADPF_U5qv64gz2QkgRzsfjNXf3lxD3_pDoZEM9E7yGatUubyNFwDqrjZ0I84Z04rdpTZI1b9NHMpyQw-1A$
>> <
>> https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/internet-history__;!!KwNVnqRv!BhF0ReQHMaqoADPF_U5qv64gz2QkgRzsfjNXf3lxD3_pDoZEM9E7yGatUubyNFwDqrjZ0I84Z04rdpTZI1b9NHMpyQw-1A$
>>> 
>>> 
>> --
>> Internet-history mailing list
>> Internet-history at elists.isoc.org
>> https://elists.isoc.org/mailman/listinfo/internet-history
>> 
> 
> 
> -- 
> *****
> Craig Partridge's email account for professional society activities and
> mailing lists.
> -- 
> Internet-history mailing list
> Internet-history at elists.isoc.org
> https://elists.isoc.org/mailman/listinfo/internet-history
>   
> -- 
> Internet-history mailing list
> Internet-history at elists.isoc.org
> https://elists.isoc.org/mailman/listinfo/internet-history
> 
> -- 
> Internet-history mailing list
> Internet-history at elists.isoc.org
> https://elists.isoc.org/mailman/listinfo/internet-history