Can Xerox's PARC find new life with SRI?
nytimes.com
nytimes.com
While there’s a need for more directed short-term research, I feel that industry has largely moved away from more free, longer-term, “pure” research, and I also feel that academic research has become affected by “publish or perish” and fundraising demands. There seems to be no room these days for researchers who want to do more exploratory work, short of becoming independently wealthy or relegating such exploration to nights and weekends.
Billions and billions (trillions?) of dollars in value was created (e.g., Adobe, 3Com, SGI→NVIDIA, Mac / Windows, laser printers, ethernet, etc.) based on their research, but ~0% of that value went to Xerox. From a business point of view, it was a disaster. And business folks are primarily the ones deciding what's funded at this level. If I'm an executive who's not sure if I, or my organization, will be able to tell the great ideas from the duds, why spend money that might expose that?
A large part of the money that was funned into research at Xerox came from the government itself in the form of tax breaks and other benefits. So it was not a big sacrifice for Xerox to maintain that lab. If they had to change course was due to government policy becoming less interested in industrial innovation and more focused on offshoring and financial gains for the last decades.
Do you have any support for that? For example, Xerox's financial disclosure forms should still be available.
Because I think it's wrong. The government funded academic projects, not commercial ones.
>Government contracts were also a big income source for Xerox, AT&T, IBM, and others.
Don't get offended if you get challenged on that. I don't think you actually know anything specific here.
High Taxes Brought Us Xerox PARC and Bell Labs [1]
PARC, a Xerox company, today announced it has secured funding from the US Department of Energy’s Advanced Research Projects Agency [2]
NTIA grants Verizon, AT&T-led open RAN group $42M [3]
Nvidia, Microsoft, Google, and others partner with US government on AI research [4]
Governor Hochul announced a $10 billion partnership with the semiconductor industry such as IBM [5]
[1] https://slate.com/business/2012/07/xerox-parc-and-bell-labs-...
[2] https://web.archive.org/web/20231211024425/https://www.parc....
[3] https://www.mobileworldlive.com/old_latest-stories/ntia-gran...
[4] https://finance.yahoo.com/news/nvidia-microsoft-google-and-o...
[5] https://www.governor.ny.gov/news/governor-hochul-announces-1...
[1] Biden to inject Intel with CHIPS fab cash 'next week':
https://www.theregister.com/2024/03/15/biden_intel_chips_fun...
The discussion was about PARC in the 70's. Not Intel now.
[2] doesn't work as a link.
In any case, all your links seem to be current, not 1970's.
Give us ANY Xerox product that was government-funded. Don't say "ARPANET" because that was not a Xerox product. Nor an IBM or Microsoft product.
As for the CHIPS Act: that seems to be fairly recent, doesn't it?
You're right. He's just throwing wild numbers around and mixing up eras.
"Dealers of Lightning" is about Xerox Parc and is also a great read if you're interested in the subject.
I don't think the level of tax breaks has really changed between then and now for R&D, though maybe reducing the headline tax rate reduced the benefit of them? But I think the main issue is that the money wormhole stopped working - Industrial research labs seem to be a kind of luxury good
You mean like google and ML and Attention is All You Need...
Or let's look at what a group of people with brains and money went on to do: the PayPal mafia. Would they not be the modern day version of att/unix(a product of anti trust)->Berkley/Unix->and the foundation of the valley (sun, most databases)...
If anything, Xerox's "willingness" (intentional or not) to allow the root ideas to grow in various places has been a blessing to the IT industry as a whole.
What you are suggesting is they could, or should, have become all the companies you listed. But that would gave been both impossible, and counter productive.
Steve jobs saw the GUI, and Ethernet, and object orientated programming, and laser printers, and office software, and yet Mac is famous for just the GUI. Others would run with networking, programming, MS with office and do on.
And in addition to those major players there were lots of others competing too.
So no I don't think Xerox management needs to be embarrassed. They saw the value printing and ran with that. They left others to develop the other ideas. And in the long run we are much richer for that.
So, from me anyway, there's no shame on them, rather a tip of the hat, and a nod to the vision they had to create PARC in the first place. They are the giants on whose shoulders we stand.
In the 1980s Apple adopted/commercialized:
GUI -> Lisa (later Mac) Desktop
LAN -> AppleTalk (NBP lookup eventually influenced mDNS)
OOP -> Object Pascal (still exists in Delphi/Lazarus); arguably HyperTalk (messages/events/handlers)
Laser Printers -> LaserWriter
Office Software -> Lisa Office System; MacWrite, MacDraw, MacPaint
And of course Microsoft Excel and PowerPoint debuted on the Mac.
But your point is well taken - PARC alumni (and ideas) did not only land at Apple; for example, most of us have probably used Ethernet, Microsoft Word, PDF (derived from PostScript), etc.
I'm not suggesting Apple didn't (ultimately) embrace those technologies, but rather that they weren't the "driving force" behind them. Whereas you could say Apple was the "driving force" of the GUI.
Microsoft would go on to drive Office software. There were word processors, and spreadsheets before Office, but Office integrated all those tools together and became the gold standard.
Adobe would drive desktop publishing, PDF and so on.
Sooo much of today's world is built on PARC, SRI, Bell Labs, and a couple others. It was an explosion of development and research seldom seen in any field outside times of war.
Dead wrong. Laser printers alone were a $2B a year business. Printers, especially big ones, were right in their wheelhouse.
If you want to argue it could have been so much more wealth, you're right.
My favorite demo they gave us was replacing the PDP-11 inside the printer, which was becoming increasingly expensive, with a Sun SPARCstation. They had done all the work to do a full demo on the prototype printer we were working on. Everything was much faster and it provided a lot more features. Plus development printer software could be developed in C or other languages, not just assembly. They were effectively told "that looks nice" and sent back to Palo Alto. That was hard to watch.
Maybe this research lab should partner with something like YC so when an idea is nearing a production quality design, VC's can fund it to product fruition with some sort of royalty agreement with the research lab to fund more ideas.
I would think a lot of highly technical, actual scientists (not some dude who's been hacking at CRUD web apps for a couple decades, aka me) would love to work at a place where constant academia/VC pressure is set aside to allow more pure research to happen.
I think Apple could've done this a few years back when they had something like $100B cash on hand - they could have established a research foundation and provided it with funding that would've lasted for long enough for it to become self sustaining. Maybe Apple still could do this as they've got $73B cash on hand. $10B would go a long ways towards establishing an institution like this. Even better if other FAANG companies joined by providing seed funding, but maybe that looks too anti-competitive these days? (the incentive for providing initial funding could be a stipulation that funding companies get a break on IP license fees)
Of course, as someone else mentions in this thread, this institution should learn from PARC's mistakes and do a better job of licensing their IP in order to keep operating for a longer term.
I suspect we really need something like this. Bell Labs (and PARC) gave our economy a technological boost that it wouldn't have had without them. Maybe a revitalized SRI (with some additional outside funding) could become what you're talking about?
Also once you start framing it as investment or a biz decision in a big public company, the natural question is whether it's the best use of capital for Apple. It's not really about short vs longterm thinking either, it's about the deployment with the highest likelihood to return on any timeframe.
At some point in the future there should be some IP that could be licensed.
If you're starting out with, say, a $10B endowment there should be investments that could sustain a certain level of operation for quite a long time until that IP license income kicks in. Towards the end of the article they talk about the land that PARC is situated on and how that's now very valuable and will provide a source of income for the new SRI/PARC - so they seem to have lucked out with a real estate investment that they didn't realize at the time would payoff later.
wasn't for example OpenAI exactly that? The difference from the old times here seems to be that any advances are easily monetized today due to gigantic VC available, and thus the immediate ROI demand immediately arises from inside - from the employees, from the CEO down to lowest rank-and-file, who see the immediate potential to turn the advance into billions on the spot. Imagine if Alan Key and his colleagues could turn their inventions into billions in the next few months after making those inventions - I doubt they would have let Jobs to take the tour and practically steal the IP.
This is part of how they avoided the ROI catastrophe (researchers with enough cred could argue that there was a long term potential) and the publish or perish issues.
2. The biggest barrier to starting a research lab is how will this lab make enough revenue to fund the lab’s operations and the living expenses of its staff? A significant challenge of running a research institute is fundraising. The first thing a lab must do is to establish a mission. Will this lab do long-term, speculative work or will it aim for shorter-term surer bets? What are the lab’s boundaries; what stipulations from funders can it live with, and what ones are unacceptable? I say this because as a researcher, I have learned The Golden Rule: “He who has the gold makes the rules.” Another way of putting it: “he who pays the piper calls the tune.” It is often the case that money comes with strings attached; it’s up to the lab to determine which strings are acceptable or not.
Because while there is a place for some hands on developers at such a place that is not most of the staff? The Manhattan project hired a lot of engineers but none of them were the design leads.
I apologize if I come off as bitter, but as someone who got into research because I was inspired by reading about people like Dennis Ritchie, Alan Kay, and many other researchers from the previous generation, it’s a shame what science has become, but it’s a reflection of our society where the worth of persons and individuals is quantified in terms of money, and where anything that’s not perceived as making money is denigrated (such as the liberal arts).
Xerox PARC did it, and reaped almost none of the rewards for their innovation, so it perished. It only worked for Bell Labs because they had monopolistic control of telephony and could charge whatever they liked to subsidize their R&D. As soon as they lost the monopoly, it went down hill fast. Google is the closest thing to this in the past couple decades, and have seen all their investments in AI walk out the door to found competitors.
Bottom line, pouring enormous amounts of money into pure research without demanding commercialization results certainly benefits researchers (who can pick up and take their research elsewhere) but runs companies into the ground.
Unfortunately the lesson industry learned from Xerox PARC is that research needs to be more tightly controlled by the company instead of the company’s direction being influenced by the research coming out of the lab, and thus industrial research labs, at least in computing, have undergone a transformation where it’s about near-term progress in areas that are closely tied to business strategy. I’ve worked in industrial research settings for much of the past decade and have seen the last vestiges of more free research die out. I can’t think of a company that was run to the ground by its research division, and even if such examples exist, this is an indictment on management for allowing research expenditures to sink the company, not research.
Of course not all companies can afford the types of labs that Xerox and AT&T had when those labs were in their golden ages. However, I believe that the shift toward research driven toward short- and medium-term business needs, as well as academia’s focus on fundraising and publications, has effectively snuffed out unfettered research, where the only people in a position to pursue research that way are grad students with appropriate advisors and funding situations, tenured professors who are willing to take risks, and those working outside the system such as independent researchers. Where’s the next big thing going to come from if businesses are worried about the next quarter’s results and if academia has effectively abandoned freedom of inquiry in a quest for more money and prestige?
The employee open house was really neat, with different labs showing off whatever they were working on, from early noise canceling tech, to computers with color screens, cell counters, you name it. I know we visited "Doug's Lab" but I have no idea what we saw there. As any aspiring nerd, I was quite impressed that my dad and him were on a first name basis.
I know government staff hates this too. When a project starts to wind down or enter the last 3-6 months, the contractors start disappearing: not showing up for meetings, progress slows, etc. because they are all out looking for their next contract so they don't have a gap. "Every day is a billable day"
https://mitmuseum.mit.edu/collections/collection/david-a-huf...
Parc, HP Labs, etc. were places of innovation back in the day - even things going on well into the 2000s.
Sadly most companies don't do or value that type of research anymore.
1. The folks running the show started out as copier salesmen who had trouble getting their heads around the fact that the banks, utilities, governments etc running high volume printers used mainframes which required mainframe software for the printers to be useful.
2. Revenue from printers (click charges) was entirely credited to sales. The software that enabled the sale and click charges was looked upon only as expense. Its contribution to revenue was ignored.
3. Technologies were abandoned at a slightly lower rate than Chromebooks. Customers were left holding the bag with dead ended platforms and applications. The concept of investment protection, providing a growth path for customer software on Xerox platforms simply did not exist.
As many have mentioned, Xerox was expert at throwing away the future and continues to shrink.
a) Having to show product market fit and appease magpie investors
b) Emerge after 10 years with devastating debt and with poor career prospects.
Those answers are what matters.
Obviously less of an issue if well paid, but well paid academia is kind of a contradiction.
"Student debt" was not a thing back in the 70's and 80's.
Secondly, PARC researchers were not paid university postdoc-level salaries. They were paid quite well. I would guess that's still true.
The latter (DEC WRL) was credited with developing the DEC Alpha CPU; DEC SRC invented Altavista which was the #1 search engine until Google arrived.
I was an Office Systems Division engineer 1984-6 and watched the exodus unfold before I left for grad school. Overall DEC SRC/WRL didn't have as much impact as PARC; there is a time and place for everything and that kind of systems research - its time had passed ..
https://www.amazon.com/Scientific-Freedom-Civilization-Donal...
What a facepalm! >_<
Making 20% profit, and freeing up that capital for other uses, is not a face palm event. Not for you or I, and not for businesses big or small.
>But many researchers who were at PARC’s halcyon early days said that its strength was that their research was unconstrained by the need to create a specific product — a notion that seems hard to imagine in today’s product-oriented Silicon Valley.
The "but" is not necessary there. Both are true. Printers became a $2B a year business.
Coulda been so much more? Absolutely.
https://livingcomputers.org/Blog/What-Really-Happened-Steve-...
Jerry Morrison and I analyzed in depth what they should have done:
https://www.albertcory.io/lets-do-have-hindsight
Most people don't know that, in the 80's, they DID spin off a number of companies, where Xerox's only investment was a license to use the technology.
Had they held onto Apple and all those, they'd be one of the biggest holding companies in the world, if not the biggest.
Are they saying that they'll raise the money by doing real estate development? Could work, they certainly don't need all of the 63 acres. Or maybe they don't need any of those 63 acres? They could sell all of it off and move SRI/PARC to some much cheaper area of the country.
https://www.nytimes.com/interactive/2024/03/26/business/sili...