[ih] Quantifying OSI

[Dave Crocker]

On 5/12/2026 1:51 PM, Bill Nowicki via Internet-history wrote:
> Marketing was saying that only a few research customers were ever going to use TCP on BSD Sockets, and those had discounts.


Around 1987, at Ungermann-Bass, we put TCP onto the 'intelligent' 
ethernet PC cards.  One of the earliest customers was Rutgers IT 
department.  Some months after they started using the product, I got a 
call from the our sales person saying they wanted to buy an IP router 
from us.

I noted to them the obvious fact that we did not have such a product and 
did not have one in the pipeline and I knew that they already knew this 
and was pretty sure so did the guy at Rutgers.

She said yeah, but he liked us. So I asked how long he was willing to 
wait.  She called back saying 3 months.

I went to the engineer who had told me he'd love to put IP onto the 
company's router and asked how long it would take him.  (He was just 
coming off a project.)  I listened while he worked it out and he said 
two months.  I said great, that's a month of testing.

Before seeking authority from our boss, I went to our marketing guy and 
told him of the opportunity.  He, too, was fully dismissive, saying that 
there was only going to be a tiny market.

My boss said we tried to never say no to a customer.  He asked about 
effort and time.  Heard the answer.  Gave the goahead.  But alas, UB 
marketing was never able to sell into a general market. But for a brief 
moment, I was competing with cisco...

d/

-- 
Dave Crocker

dhc at dcrocker.net
bluesky: @dcrocker.bsky.social
mast: @dcrocker at mastodon.social
+1.408.329.0791

Volunteer, Silicon Valley Chapter
Northern California Coastal Region
Information & Planning Coordinator
American Red Cross
dave.crocker2 at redcross.org