Morris Tanenbaum, inventor of the silicon transistor, has died

https://spectrum.ieee.org/in-memoriam-may-2023

by furcyd • 3 years ago

470 100 3 years ago

spectrum.ieee.org

Morris Tanenbaum (American, physical chemist / transistor researcher) [1], not to be confused with Andrew Tanenbaum (American-Dutch, CS professor who was responsible for MINIX) [2]. Both had great significance in what we call "computers" today, in their respective fields.

[1] https://en.wikipedia.org/wiki/Morris_Tanenbaum

[2] https://en.wikipedia.org/wiki/Andrew_S._Tanenbaum

rvz• 3 years ago

Exactly. and deserves a black bar.

Can we please have a black bar for respect and recognising his great contributions?

reasonabl_human• 3 years ago

Black bar?

FeistySkink3• 3 years ago

Andrew Tanenbaum is in a no way Dutch. States as much in his Wikipedia article. There are plenty of Dutch scientists, including in CS, so no need to gekoloniseren another one.

tgv2• 3 years ago

He's lived and worked in The Netherlands since over 35 years, possibly 40.

marcodiego• 3 years ago

Sorry: https://en.wikipedia.org/wiki/Wikipedia:Lamest_edit_wars#Eth...

jakjak123• 3 years ago

Hah, I read many books by Andrew Tanenbaum. I remember him putting in dry comments like 'It is not known whether CPUs have dreams during their deepest sleep states.'

kkarakk• 3 years ago

i loved those comments, felt like a mentor talking to you instead of a person passing on info

shever73• 3 years ago

Thanks for the clarification, I was wondering if they were related.

woodruffw4• 3 years ago

It's incredibly humbling to be reminded how young this field of our is: this site and everything its presence implies (about the Internet, global communication, computers) would not exist were it not for the work of a very-recently-deceased engineer.

Very few other fields (aerospace and modern pharmacology, among them) can say the same thing.

cubefox2• 3 years ago

It is interesting to think about about alternative history here: If microchip technology had never been invented for some reason, what would we have developed instead? What is the next best computing technology?

woodruffw4• 3 years ago

I think we would have ended up here anyways: to assume that the silicon transistor would not have been invented without Morris Tanenbaum is to succumb to the Great Man theory[1].

The reality is that Morris was a brilliant man whose life intersected in the right place and time with the arc of history; we should celebrate his life and contributions for that, and not because we think that we'd be stuck in 1950 without them.

[1] https://en.wikipedia.org/wiki/Great_man_theory

Cthulhu_• 3 years ago

Vacuum tubes was the thing, I'd imagine there would be more of that.

That, or mentats.

piersj225• 3 years ago

This is something the Russians worked on,

https://en.wikipedia.org/wiki/Water_integrator

I've been told analogue computers got pretty far too,

https://en.wikipedia.org/wiki/Analog_computer

petra• 3 years ago

Possibilities:

Mems vacuum tubes.

Nano electromechanical relays: https://en.wikipedia.org/wiki/Nanoelectromechanical_relay

sawyna• 3 years ago

I don't think anyone has ever explored an alternative here. Transistors came out of exploratory research that Bell Labs funded and as the other comments say, this indeed happened very recently! I suppose the silicon industry and the digital age built up so fast that we didn't really have the incentive to explore the alternative?

kens• 3 years ago

Nowadays, people think of just vacuum tubes and then transistors, but in the 1960s there were a lot of different technologies being used. For example, superconducting cryotrons, magnetic core logic, parametric-phase-locked-oscillator logic, microwave logic, electroluminescent-photoconductor logic, and tunnel diodes. If transistors didn't pan out, there were lots of other technologies that could have taken over. Magnetic core logic (not to be confused with magnetic core storage) in particular was used in several computers and seemed poised for success until advances in transistors killed it off.

More: https://spectrum.ieee.org/the-vacuum-tubes-forgotten-rival

smaddox5• 3 years ago

Interesting hypothetical. If semiconductors were never discovered, I suspect we would have kept shrinking vacuum tubes, eventually down to the point (which has now been demonstrated) that atmospheric pressure is sufficiently low pressure to operate at.

cubefox2• 3 years ago

I'm embarrassed to admit that this is the first time I hear of him.

irrational• 3 years ago

You are one of the lucky 10,000 today.

supernova87a• 3 years ago

Agreed, odd that the name just is not in the same group of people as Moore, Fairchild, Noyce, etc? This is the first I had heard the name.

hutzlibu3• 3 years ago

Fame is never fair and balanced, it seems. I would think there are lots of other engineers, who don't even have a wikipedia article, who made important contributions as well.

rchiang• 3 years ago

Even prior to those guys, Shockley, Bardeen, and Brattain have the 1956 Nobel Prize in Physics for their work. Kilby received the same in 2000 (Noyce passed away in 1990).

newswasboring3• 3 years ago

This is the same fallacy as when people say Europe doesn't contribute to tech industry because they don't have huge software companies. All the software companies progress on the back of several asian and European companies. Most semiconductor manufacturing equipment is made either by European companies (ASML, Zeiss etc.), or by asian companies (TEL, TSMC, Gigaphoton etc.) This is why ASML till a few years ago used to bill itself as "the most important company you have never heard of".

lordfrito• 3 years ago

This is a bit late, article (and wikipedia) says he died on February 26. [1]

[1] https://en.wikipedia.org/wiki/Morris_Tanenbaum

jchallis• 3 years ago

It's incredibly humbling to think that his #2 and #3 life achievements were building practical high field superconducting magnets (making MRI machines possible a decade later) and being CEO of AT&T.

dakial1• 3 years ago

"Despite Tanenbaum’s early work on silicon transistors, AT&T did not support further research or advancement of the technology. At the time, Bell Labs was the research arm of AT&T. Although Bell Labs had “a significant technological lead in silicon transistor technology, it stopped doing proper research in the field—partly because it just wasn’t immediately relevant to AT&T’s business—so silicon transistor technology, including the integrated circuit, was done by Intel and Texas Instruments instead,” Tanenbaum said in a 1999 oral history conducted by the IEEE History Center."

Good call AT&T! /s

Turing_Machine• 3 years ago

so silicon transistor technology, including the integrated circuit, was done by Intel and Texas Instruments instead

Right. Tanenbaum invented the silicon transistor (which of course was a major accomplishment -- definitely worthy of honor), but what most people think of as a "microchip" (i.e., an integrated circuit with multiple components on the same piece of semiconductor) was invented almost simultaneously by Jack Kilby at Texas Instruments and Robert Noyce (then at Fairchild, later a cofounder of Intel).

Fatnino2• 3 years ago

I daily drive past the spot where Fairchild was at the time. It's next to the cheapest gas station in Palo Alto (not really saying much) and across the street from the location that used to be Sun's HQ before they moved to where Facebook is now.

allturtles• 3 years ago

Here's another source that disagrees (BTL = Bell Telephone Laboratories):

"In spite of its secondary role as a semiconductor supplier, the course of semiconductor research was shaped by the scientific and engineering achievements at BTL during the 1950s. While most of the semiconductor industry concentrated on manufacturing germanium transistors and diodes during this decade, BTL spent most of its research dollars on silicon devices... Of the ten major events instrumental in the demise of germanium as the preeminent starting material, seven originated at BTL while the other three were derived from BTL research activities. "[0]

[0] https://ethw.org/w/images/3/3a/Seidenberg%2C_1997.pdf

mlyle• 3 years ago

That -agrees-. AT&T built a substantial lead in silicon devices during the 1950s, like your source says.

TI and Fairchild pulled firmly ahead in the 1960s; Intel became hugely relevant in the 70s and 80s. AT&T had mostly abandoned silicon research by this time.

andyjohnson0• 3 years ago

The tree of technological and scientific developments that enable almost all of us - on this site - to do what we do, narrows back to just a few people and their work. It seems that this man was one of them. Thank you Mr Tannenbaum.

Black bar.

MichaelZuo• 3 years ago

It seems the headline is a bit editorialized with 'Microchip' as he does not seem to have been involved with subsequent developments of the integrated circuit, what microchips commonly refer to.

Though, after checking, there apparently are no clear cut 'inventors' of the integrated circuit, which strikes me as a bit odd.

dang• 3 years ago

The article seems to imply that "silicon transistor" would be more precise than "microchip" so I've put that in the title now, at least until someone explains how it's wrong.

deepnotderp• 3 years ago

In so far as the modern monolithic integrated circuit is concerned, it’s pretty indisputably Bob Noyce imo

pclmulqdq• 3 years ago

Kilby from TI is often credited as inventing the integrated circuit. But so is Noyce. And a few others. I think Kilby is actually the correct answer.

joshxyz• 3 years ago

rest in peace sir, wishing you well in your journey.

jschveibinz• 3 years ago

If you haven’t read “The Idea Factory” by Gertner, it’s really worth the time. The book is well written and it covers a very exciting time in our technological progress. Dr. Tanenbaum is discussed in the book along with the rest of the dream team at Bell Labs during those years.

Here is a book review if you are interested:

https://www.dau.edu/library/professional-reading-program/Pag...

ed25519FUUU• 3 years ago

94. Well done, sir. Enjoy your rest.

intrasight5• 3 years ago

I went looking for a recent estimate of the number of transistors that have been manufactured. It appears to be in the range of 10^23. I'm pretty sure there in nothing else anyone has invented that has been replicated in such volume.

analog31• 3 years ago

Wow, nearly a mole. I remember thinking that a mole of any "thing" was impossible.

taneq• 3 years ago

1/6th of the way to Avocado's Number. ;)

intrasight5• 3 years ago

https://spectrum.ieee.org/transistor-density

Interpolate out and we'll be at a mole in 1-2 years

colordrops• 3 years ago

What about bits on drive platters? Do they count as individually manufactured units?

tomnipotent• 3 years ago

For a modern SSD, it requires several transistors to store a byte so it's still transistors all the way down.

irjustin2• 3 years ago

I'll be the first to admit this is a meaningless debate but seems fun!

I'll say no

We don't colloquial think of a HDD bit as a manufactured unit on magnetic disks. It doesn't have a direct macro connection/object.

For SSDs the macro analog of a bit storage is a register and those are like 6 nand gates which those are made of like 3 transistors?

Exact numbers are wrong I'm sure. It's been a long time.

djmips2• 3 years ago

You can still walk into a store and buy one transistor but you can't buy one storage bit. Not sure if that is an argument but interesting.

dghughes• 3 years ago

It's amazing to think of a thing someone has made or has helped make be so numerous. And a physical thing too not software or words or an idea an actual thing.

Transistors on a CPU are actual things. Intel or AMD CPUs having billions. One CPU the Cerebras Wafer-Scale Engine (WSE) as they describe it "has 2.6 trillion transistors, which make up 850,000 AI-optimized processing units". That's one big chip!

hovden• 3 years ago

Interesting. Do you have a source?

How does it compare to other tech, like nails? Or cups? Shoes?

Fatnino2• 3 years ago

Letters?

bnegreve• 3 years ago

Be careful with large numbers: 10^23 shoes would be roughly 854B shoes for every (~117B) human that have ever lived, that is about 14 million pairs of shoes per individual and per day. I am pretty sure there has never been a market for that many shoes.

intrasight5• 3 years ago

IEEE

https://spectrum.ieee.org/transistor-density

Interpolate out and we're close to 10^23

devsda• 3 years ago

Interesting numbers.

How does it fare if we compare against those that are measured by weight ?

Edit:typo

streakfix2• 3 years ago

Penicillin molecules

bibanez• 3 years ago

I was thinking the same! Something bio, that has the benefit of exponential growth

intrasight5• 3 years ago

Discovered not invented. And grown not fabricated.

fsckboy• 3 years ago

in the range of 10^23. I'm pretty sure there in nothing else anyone has invented that has been replicated in such volume

Covid-19 coronavirus particle/strands

jkuria• 3 years ago

where's the black bar on HN @dan?

newswasboring3• 3 years ago

I second the call for a black bar. We wouldn't have any of this (gestures vaguely at everything) without this dude.

aarobot• 3 years ago

Apparently he passed away on February 26th, so the black bar could be misleading.

stevefan1999• 3 years ago

Is he by any means related to Andrew S. Tanenbaum, the author of Minix?

gpvos2• 3 years ago

"By any means" yes, we all descended from a common ancestor.

osigurdson• 3 years ago

> it stopped doing proper research in the field—partly because it just wasn’t immediately relevant to AT&T’s business

It feels like businesses today are better at pivoting. Imagine improving the state of the art by 10X and simply shelving it.

eternauta3k• 3 years ago

Maybe pivoting AT&T is a bit like pivoting a planet-sized supertanker.

pyro123• 3 years ago

Is he related to Andrew S Tanenbaum or is it a common second name?

gpvos2• 3 years ago

It's fairly common. May indicate Jewish descent. https://en.wikipedia.org/wiki/Tanenbaum

1letterunixname• 3 years ago

Dern. I had a physics lecturer, of all places, DeAnza College (Cupertino, CA), who worked in that lab. OG engineer teaching physics/engineering fundamentals.

Morris Tanenbaum, inventor of the silicon transistor, has died

Related Stories