Back in the old days, companies like Dupont or Shell would have research groups that had massive lattitude in what they could work on. Not all groups in the company by any means, but there would be some that did pure curiosity-driven research, broadly in the remit of the company’s interests. Bell labs is another famously innovative company that had research groups in this vein.
My understanding is that these groups have all gone - short term shareholder value is king and cannot and does not support this sort of longer term research vision. Obviously companies still do fundamental R+D, but it is focussed on the immediate term.
Anyone know of any exceptions left? I was thinking maybe software / computer science might lend itself to this sort of research. Do google and their ilk have groups who’s remit is to basically dick around until they find something interesting?
From Wikipedia, “As a motivation technique, Google uses a policy often called Innovation Time Off, where Google engineers are encouraged to spend 20% of their work time on projects that interest them. Some of Google’s newer services, such as Gmail, Google News, Orkut, and AdSense originated from these independent endeavors.”
I understand that google lets engineers work 1 day a week on their own projects and a number of well known google applications came from that “20% time.”
I am in such a group, but as with academics, we are largely dependent on funding. Of course, so were many of those older groups. We also occasionally try and partner with those bigger companies on getting funding, so I know they still are doing research.
It is absolutely true though that short term profit based investment has hurt research. Funding is much harder to get these days and research is constantly on the firing line. Accountants don’t like to pay large amounts of money for things that may or may not produce a profit in ten years.
Intel Corporation funds some research that is not directly tied to core compencies, along with the long term research related to developing manufatcuring and leading edge technologies. It also has a venture capital arm that may be considered investment in future, unrelated research.
I used to work at Bose headquarters and it runs lots of long-term research projects, some of which have little to do with sound. It is a private company and set up to stay that way so it doesn’t have to worry about quarterly results much.
Bell Labs did a lot of fundamental research, a lot of its people did fundamental research with limited application. In return, it produced umpteen Noel prizes, discoverd big bang background radiation, developed UNIX, C language, transistors and CCD image sensors, among far too many accomplishments. Of course, it was financed by a hidden tax on monopolistic inefficiency, the phone company.
The Large Hadron Collider? That’s all I can think of. I have a friend who does post-doc work at UC-Berkeley in Materials Science/EE, and a lot of Materials Science is pretty fundamental research-y. Universities, especially state-run ones like UC-Berkeley and MIT, are probably the last bastions of fundamental research. DARPA comes to mind, too.
The 1990s saw the demise of lots of corporate labs, either because the company itself died, or they thought that they couldn’t afford it. That’s when the big Bell Labs in Holmdel closed (although there are still Bell Labs open, even in the US.) That was also the time of closing of
Polaroid
AOtec Labs (the last of American Optical’s R&D efforts)
GTE Labs in Waltham, MA
Textron Defense Systems in Everett MA (the former Avco Everett Research Labs)
There is a definite move to have most physical R&D done in University settings, rather than in companies. Entities such as Lincoln Labs are still around (associated with MIT) and CREOL (asociated with UCF), but Kodak isn’t likely to be doing any corporate research into digitial imaging in the future.
As I said, Bell Labs was the classic example. FOr decades, Bell/AT&T enjoyed a monopoly on teh phone system. You could not attach an extension without paying them to do it for you. Long distance cost a lot. There were operators, phone books, house call techs, bureacracy galore and much more all paid for by that overpriced phone service. Technology and anti-trust destroyed that cozy setup and the gravy train of limitless cash. IIRC, 50,000 middle managers for example were laid off during the decade that the Bell breakup was implemented and the phone system deregulated. The creator of Dilbert worked for PacBell, IIRC, and drew his inspiration for aimless clueess bureaucracy from their workings during that time.
many others - Kodak, IBM, Polaroid, 3M, Xerox - all sold their products with a massive premium based on reputation, and used that extra to fund their research institutes. During the 80’s and 90’s, massive cost cutting and competition in what used to be near-monopolies destroyed the gravy train that allowed this sort of luxury. Lean and mean means no researc that does not have a short term payoff.
The equivalent companies today - Cisco, Google, etc. - may not have the longevity to make their mark in the volatile business of high tech but at least they are tring.
Yeah, and Stanford does some fundamental research too, from what I understand. Private universities rarely have the endowments to really compete in this area, though, I’d say.
Academic research is funded by external grants, not endowments. In fact, the research program is often a “profit” center, since the host institution takes a cut of all grants awarded to its faculty.
Which basically means there’s not much difference between public and private university research, taken as a whole.
A few weeks ago, Ira Flatow, the host of the NPR program Science Friday, interviewed the author of a book about Bell Labs called “The Idea Factory: Bell Labs and the Great Age of American Innovation.”
According to chemists I’ve known who have worked for or interviewed with 3M and Dupont, there is still some mechanism for this type of research built in around other tasks. Many pharmaceutical companies have small R&D departments doing novel drug design and discovery, as well as developing new dosage forms and drug delivery methods. These employees are usually ones with several post-doctoral certificates after having spent a short lifetime studying things like synthetic organic chemistry.
This is not long term research by any stretch of the imagination, but rather a good way of coming up with short term new applications.
Good long term research is not targeted to just improving existing product lines, but possibly to open up totally new areas. Microsoft would seem to qualify. IBM still has some, but even 15 years ago some IBM research people were getting contracts from product groups. Most people I know of at Intel Research are tightly focused - if the management was really doing a good job, they might have figured out the future was in phones and pads and not be losing so much market share to ARM.
I think DuPont still does some. But there isn’t a lot anymore.