Lowood and Berlin

We met, all thirty or so of us, in the Texas Instruments' Sunnyvale auditorium (which can seat four hundred) on a rainy evening the second of December, 2014.   

To open the December 2014 meeting, Alan J. Weissberger, chairman of the IEEE SV (Silicon Valley) Tech History Committee, posed the question to Leslie Berlin and Henry Lowood, two of Stanford’s esteemed historians of technology: “What is the process, method and procedure for doing authentic, historical technology research?” 

Weissberger went on, though, to the observation that “Bad books, repeating obvious flawed stories, keep appearing.  What can the Stanford Archives project do to reduce these?”

What an opening for Lowood!    He seized the moment grandly, saying, “Bad books perhaps are a good thing—they generate more research and provide impetus to create good books.”

That riposte stopped the group, for a while.  I couldn’t stop thinking about it overnight.

Essentially Lowood gave voice to the same point that Jennifer Light made at Dearborn--there are too few of "us" and too many of "them" and there is little scaling capability at the moment in this field.

Berlin's book, "The man behind the microchip: Robert Noyce and the invention of Silicon Valley," drew appropriate praise.  It was a ten year (no, make that fifteen, the way she expressed it, growing out of her PhD thesis) project, and remains the only decent biography of the man.  There is no biography (yet) about his partner, Gordon Moore (think Moore's Law, the most important legacy in many respects of the integrated circuit world that has given us incalculable impact on society).  Arnold Thackray and David Brock are finishing their tome about Moore at the moment; they've been working on it (with full funding) for a decade.  

The snide asides, from Weissberger and several others, were about Mike Malone's The Intel Trinity, which emerged in August.  Some fifty Amazon reviews rate it 4.5 out of 5.0, very solid for such work.  Several reviews, though, slam it almost unmercifully for its 'laziness' or inaccuracies or blatant hucksterism.  A few take it on for its lack of solid or even correct data.  Emotions, amongst the aficionados, quickly run amok.

I know Michael, pretty well.  He wrote "Bill 'n Dave" about HP's founders around the time we wrote "The HP Phenomenon."   It was well written, as is The Intel Trinity.  Mike can write engaging, emotional, smooth prose--he's a novelist masquerading as a historian, what many call 'semi-fictional history.'  He's written some 25 books, mostly about Silicon Valley antics and folk.  We are much the richer for having his pen and his insight.  But, and he would probably agree, don't treat it as totally factual.  It is Internet-researched, almost never based on primary sources.  And much of it--in every book--is anecdotal, mythical, unsubstantiated.  He's a perfect foil for the question asked of Lowood.

But Lowood's answer was the surprise of the evening.  I've mostly come around to that view myself.  It's one way to rationalize the popularity of such works as Walt Isaacson's Steve Jobs, which is beautiful writing, about a fascinating individual, where the author believes himself that it is history when it is not--it is biography, and gets way too many things factually wrong about Apple and about Jobs' own actions.  And, no one has incentive to go 'right the wrongs'--it wouldn't get read, for one, and it wouldn't sell, for two, and maybe we could argue even that it doesn't matter.  But of course, it does matter.  For ethics, for truth, for trust in our histories, it matters.

HP and its 'historian' still cling to the wrong idea, prominent on their website, that HP started with selling eight oscillators to Disney, for $71.40 each.  Great start.  Wrong on all three factual counts.  Disney was not the first customer (Stanford and Caltech were); Disney bought nine, not eight, units; and they paid $57.50 per unit.  Primary sources reveal this; HP's own archives 'knew' this for 25 years, and then 'magically' it got changed.  Showing the facts to HP a few years back drew the threat of a lawsuit to stop our book--great support of truth-seeking.  Is there percentage in trying to 'right this wrong' at this point?  I think not--its like believing with Pandora that the "ills" can be recaptured, once loosed on the world.

This whole essay ducks the question of plagiarism--the Doris Kearns Goodwin and Stephen Ambrose frauds still stick in my craw--made even worse because such "worthies" as Walt Isaacson can't wait to leap to their defense while they are being 'initially' skewered (sort of like Oprah jumping to defend Bill Cosby?).  

Is there an answer?  Not while historians are paid a pittance, and good novels sell in the millions.

SIG/CIS at Dearborn

Who does history?  Who should do history?  Why do we (they) do it?  How are the choices of what to study made?  Who vets the resulting work? 

These questions provoke widely varying points of view. 

Just prior to the annual SHOT (Society for the History of Technology) conference at Dearborn, Michigan in November, the Special Interest Group for Computers In Society (SIG/CIS) had a spirited e-mail exchange on this topic.  The topic centered on the perceived historical flaws in Walter Isaacson’s newly published book, The Innovators: How a Group of Inventors, Hackers, Geniuses and Geeks Created the Digital Revolution, and what should/could be or have been done.  Views varied widely, from solid enthusiasm for a truly well-written, engaging book about “our topic” to real concern for the lack of involving the SIG/CIS leadership as reviewers prior to publication—essentially vetting the book.

The keynote speaker at the SIG/CIS meeting, Jennifer Light, tackled the question of “who” head-on, averring that there are (at least) five different communities who are focused on the history of computing:

1.     Historians of technology (us, in that room, as SHOT and SIG/CIS members)

2.     Historians of science (them, having their annual conference the same week in Chicago)

3.     Practitioners in the computing design worlds (architecture, computer-aided tools, etc.)

4.     Practitioners in the communications/media worlds (movies, games, iPhones, etc.)

5.     Journalists, analysts, and amateur historians of many stripes and origins

Light’s key observation, for my taste, was that none of these five groups talk to each other.   Bernard Carlson, professor and Chair of the Engineering and Society department at the University of Virginia, lent special emphasis to this point when he mentioned that the head of the National Academy of Engineering  (unnamed at the conference, but it was Bill Wulf) asked to get the SHOT periodicals, but after a year of reading them, he reported to Carlson that “I don’t understand anything they’re talking of.”

Help me, please!  The chairperson of NAE for America, the premier group of eight thousand engineers in America, himself a pedigreed PhD with long engineering experience, numerous patents, business success and accolades, could not relate to the issues, debates, or even vocabulary of the leading historians of technology of the world?  It’d be hard to make up this story, it sounds so ludicrous. 

Light continued, noting that at most, there are perhaps four hundred practicing Historians of Computing Technology in the world today, versus an observation by the most popular author in that field (Jim Cortada) who recently wrote that “we will come to see this field as the most pivotal historical point in civilization.”

Cortada has a superb little book about 'our craft'--"History Hunting"--wherein he notes: "in the case of the world of computers, we have had the problem that so few participants have been interviewed and so many thousands could be.  Computing pioneers also have had the bad habit of dying, even if many have lived into their eighties and nineties.  Because doing high-grade interviews is so expensive and time-consuming, we needed to find another way to get that task done if we were going to scale up to interviewing many people rather than just staying content with a few.  Remember, there are only several hundred professionally trained historians in the world looking at the history of computing, and the number of people they can interview is low....   (After CBI, CHM and the Software History Center collaborated, got a $600K grant from Sloan, and built a website, sought lots of folk, etc....) the project proved to be a partial success."  And that overstates the success, measured against the original goal.

So... an opportunity awaits!  Or maybe multiple opportunities.  Historians are too few, amateurs are too amateurish, practitioners are too self-absorbed, etc.  But... meanwhile we're losing valuable stuff.

Capturing History

We've posted before in this blog about the importance of "Capturing History" for our High-Tech community.  See, for example, "An Added Focus for InnovaScapes Institute" posted 1/7/14, followed by "What have we learned via Oral and Video Histories" posted 7/19/14, and the sequel ""Why Collect Oral and Video Histories?"" on the same day.

Two recent events are worth noting in this regard:

First, the annual SHOT and SIG/CIS meeting in Dearborn, MI the second week in November

Then, the IEEE Silicon Valley chapter December 2014 meeting at the Texas Inst HQ in Sunnyvale.

The next two posts will report briefly on these events.  What this post does is introduce the topic that they have stimulated for me:

"History is too important to leave up to historians."

Seriously, we have a huge problem in this regard, and almost no one is aware of it, or more importantly, how they can help influence it.

For example, Christophe Lecuyer, an energetic, focused historical researcher long associated with Silicon Valley history recently was able to publish a small volume of "the Fairchild papers."  These are key notebook entries from the folk who famously bolted from Shockley Labs, pioneered almost all of the key technologies for silicon chip manufacture, and then created Intel Corporation.  It is a wonderful glimpse at "How it Happened"--where "It" is the crux of the digital revolution that will go down in history as probably THE epochal event for civilization's knowledge management.

The story, though, worth reporting is how closely we came to losing all of this historical record.

Fairchild Semiconductor eventually had more folk leave to start new companies than anything experienced at Shockley.  Intel, AMD, and National all spun directly out; Zilog was part of the second wave, etc.  Within a decade, more than 40 semiconductor companies were competing in what was just being termed "Silicon Valley."  Another decade, and most of them had disappeared, merged, or struggled against the Japanese 'onslaught.'  In fact, the irony is that Silicon Valley was no longer the home of much silicon by the time (1982) that National Geographic did its spotlight article re SV, and the Japanese had decimated the local firms by producing much higher-quality parts (as opposed to the dumping charges that U.S. manufacturers "manufactured").

Fairchild itself struggled, then was sold, first to Schlumberger, and then to National, its historical records boxed up and retained, but not examined at all.  Then National, years later, got sold to Texas Instruments, which had someone discover the historical records, start to dispose of them, and then thought to call the Computer History Museum to ask if they'd be interested in the corpus.

Wow!  What if they hadn't called, but instead dumped the records?  It urns out that we know most of the records of most of the companies in Silicon Valley history have been discarded, shredded, lost.  Irretrievably.

And we are losing the pioneers.  People are mortal, and our heroes are no exception.  So the leaders, the inventors, the creators--never mind the legions of helpers--are seldom recorded for their history, their experience, their perspective.

Does it matter?  We think so.  Can we got a grip on this?  How?  There are only an estimated 400 professional computer historians in the world--see the next post--hardly enough to do justice to the field reworking all of the world's societal infrastructure.  Moreover, most of those 400 are focused on the minutiae of a few people or companies, or even one or two specific machines (e.g. IBM 1401).

So--tell your friends, tell a CEO, tell your neighbors--SAVE THOSE RECORDS.  Preserving is the first step toward later interpretation.  Without it, nothing can easily be done later.