I met with Nat and Bob at Nat's home in Oxford, Connecticut. Nat was open and welcoming but asked that I not use a tape recorder during our conversation though he had no objection to me taking notes. Following are the notes I took during our conversation that afternoon, retyped for presentation at VCF 7, November 2004.
SCELBI was two years old when BYTE started; Wayne Green used the SCELBI mailing list of 12,000 names and got a 67% response to his mailing.
early interest in electronics, ham radio operator at 14- early business assembling Heathkits for other
high school dropout, joined Navy and got HS equivalency- night school after Navy for additional schooling
interest in computers from 1965 - electro-mechanical experience at Bunker Ramo
1968- PDP/5, PDP/8
1971- BSEE (from University of Connecticut), met Bob Findley on campus got job with General DataComm, bought his own PDP-8 and teletype
got the idea of making a "personal computer" for engineers; felt there was a market for computers for individual rather than for companies
original target markets at time of design (1972) - (1) tool for engineers, (2) applications for handicapped
late 1972 - attended Intel 8008 seminar at GDC - saw potential applications for both GDC and personal computer - GDC management not interested in in-house applications
NW approached Findley and one other employee to chip in and buy parts in order to build three computers for themselves
designed breadboard systems; ordered 100 boards, drilled holes by hand
co-designed with others - emphasis on user interface and system segmentation
8008 cross assembler for PDP/8 written during prototype development
approximately $1,000 pooled to build first units; CPU cost ~ $120
July 4, 1973- first power-on of prototype; running OK by July 7
interest in commercial application pretty much from the start
August, 1973 - incorporated as SCELBI Computer Consulting (SCientific ELectronic Biological Computer) - quit GDC to pursue business full time, started commercial production of cards at this time
outside investors interested early on- the SCELBI $500 compared favorably with 4- function HP35 @ $395- wrote an HP35 equivalent math package for the SCELBI
November 3, 1973 - had a severe heart attack at age 29 - laid up for 6-7 weeks, investor interest dried up; NW was SCELBI's only full time employee
continued to work on first production run of 10/20 board sets (~ 100 boards)
January, 1974 - first unit sold to interested investor
March, 1974 - first ad in QST Magazine
April, 1974- Bob Findley became full-time employee number 2, started to hire staff
May, 1974- NW had second heart attack; ads already in print- 1,000 responses (bingo cards and letters) per week
open heart surgery; laid up from May until September while other systems came to market (Titus, Martin, Wintech)
deserted by investor, BF working alone
"Machine Language Programming for the 8008" written while in the hospital; sold 1,500 copies-also sold many copies of the SCELBI manual-book revenues kept the company running-later figured out that SCELBI lost ~ $500/machine sold
8-H (4K) upgraded to 8-B (16K) with ROM resident editor/assembler/debugger
early machines to engineers, other microcomputer companies (one to O hio Scientific)
desire for quality-parts tested in in-house test fixtures, assembled-and-tested machines burned in; also sold bare boards and kits
devastated by appearance of Altair-couldn't understand how it could sell at that price
January, February 1975-considered development of 8080 machine but decided they couldn't compete with MITS -large cancellation of orders with appearance of Altair early 1975- decided to sell off inventory, get out of hardware business-switch to books and software
1975-78-publication of Software Gourmet guide book series
major accomplishments-all prior to appearance ofMITS Altair- reliable tape interface (most difficult part of design), alphanumeric oscilloscope video display (8 x 22 characters), TTY interface, plans for disk interface
technical accomplishments not unique but distinguished by focusing on individual affordability and utility-company vision was HP quality with Heathkit pricing
comprehensive software -only 8008 BASIC (SCELBAL)
bid to Heathkit to do software development not accepted
published all source with SCELBAL (4-5,000 copies of manual sold)
December, 1976-BF left the company
publishing continued for several years-sold business to Hayden in late 1982 -sold approximately 5.5 million books before sale to Hayden
March, 1983 - NW had second heart surgery-"restored to life"
At the time of this interview (one year later), NW was planning a new magazine focused on pocket computers
same early interest in electronics as Nat -learned electronics in school starting at age 12
BSEE in 1971 from UConn- Nat was his classmate; vowed at graduation to avoid Computer Science
job at GDC as a test engineer-discrete logic, digital design-appreciation for microprocessors
switched job late 1972- still at GDC but as a programmer
SCELBI development- prototypes wired with bell wire; first interface was TTY started work on oscilloscope interface
[Nat interjected here-they decided to hold advertising until oscilloscope interface and cassette interface were working so that they could offer a complete system]
display was 22 char by 8 or 16 lines; required 2 MHz bandwidth scope
Bob wrote drivers for scope display-constant refresh; keyboard interface added to scope
ROM-based software (monitor/editor/debugger) allowed TTY or scope I/0
started working in Nat's house in Milford; soon after NW had first heart attack and was hospitalized; BF took over entire operation including answering mail, buying components, assembling first systems, etc.- many 12+ hour days
second office set up at Milford Post Road address- Nat's PDP/8 was the office computer for mailing lists, catalogs, etc. - BF did carpentry for new office too
reworked Machine Language Programming book for 8080 and 6800, added floating point package, BCD arithmetic, general purpose subroutines - sorts, I/0 processing, etc.
Mark Arnold developed SCELBAL independently with Nat (Mark lived in Wyoming at the time)
growth of "hobbyist" personal computer was slow because best engineers and designers could always get more in industry
sold "full up" system to IBM in 1974
NW: "Understanding Microcomputers" (published 1974) introduced concept of data storage as Post Office boxes.
money lost on hardware recovered through software sales
contribution to industry not recognized
championed by Steve Gray