5 ms·

That Elizabeth Warren tweet mentioned in the article stands out to me. She has spoken about breaking up big tech in her campaign as well. While I completely ag

That Elizabeth Warren tweet mentioned in the article stands out to me. She has spoken about breaking up big tech in her campaign as well. While I completely agree with regulation ensuring fairer competition; as we are seeing how much abuse total control over a computing platform can lead to in Apple vs Epic case, is breaking them up the right way? R&D is costly, and these companies spend a lot on it which is the reason US is still the leader in a lot of tech areas. The ability to spend that much just comes from their size. Breaking up companies might lead to inability of companies to go for moonshots and invest in cutting edge stuff. With China catching up and it being unlikely of them to restrict their companies this way, will it reduce competitiveness?

istorical 6y ago
creating new regulations on tech firms that make it more difficult for them to stop new competitors from arising or from protecting their monopolies makes so much more sense than just artificially deciding to slice up our top performers in industry and hamstring the sector.
- RandomTisk 6y ago
  That's assuming the monolith is actually doing good for society. I for one am desperate for more options for email providers with actual spam filtration, phones without anti-consumer spyware and eco system lockdowns, and social media that doesn't force users to speak only of things the DNC would approve of.
  - Nasrudith 6y ago
    Haven't we learned from the death of Socrates that "good/bad for society" is a terrible and meaningless metric as it is so values dependent? All of those things already exist by the way - Google ironically can help you find it but so can most other search engines. There are plenty of paid providers in addition to roll your own and many anti-spam solutions. Fairphone/Freephone or even just some unlocked down Androids will run whatever you want and can, federated social media like Mastodon has all sorts of stuff that wouldn't be approved of by the mainstream like niche pornography.
  - SpicyLemonZest 6y ago
    You're glossing over all the good things they've done to lay the groundwork for what you want. I'd like a smartphone with better privacy too, but I wouldn't want a smartphone at all if not for tech industry innovation over the past decade. I would (and did) choose a compact flip phone in a world where Blackberry was state of the art.
JumpCrisscross 6y ago
> R&D is costly, and these companies spend a lot on it To what end? And how efficiently? If there is one lesson from the last millennium, it's that innovation has poor economies of scale and is stifled by centralisation. Capital markets have not been more liquid in our lifetimes. In 2018 alone, venture capital "invested $251 billion," [1] representing the median lifetime earnings of almost 100,000 college graduates [2][3]. "25 percent of all VC deal volume" was in $1 billion+ deals, the median lifetime earnings of over 160,000 high school graduates [4]. Google is no SpaceX or Tesla or Moderna Therapeutics. It has its vanity projects. But those are run in start-up analogs. Google brings no advantage beyond its checkbook, and has a track record of its projects outperforming as spin-outs its internal operations. [1] https://www.mckinsey.com/industries/private-equity-and-principal-investors/our-insights/mckinseys-private-markets-annual-review https://www.mckinsey.com/industries/private-equity-and-princ... Venture capital’s very good year [2] https://cew.georgetown.edu/wp-content/uploads/2014/11/collegepayoff-complete.pdf https://cew.georgetown.edu/wp-content/uploads/2014/11/colleg... Figure 1, page 3 [3] $251 billion raised in 2018 / ($2,268,000 Median Bachelor's Degree Lifetime Earnings in 2009 dollars x 1.17 Inflation Adjustment to 2018 dollars per [a]) = 94,590 [a] https://www.usinflationcalculator.com https://www.usinflationcalculator.com [4] $251 billion raised in 2018 / ($1,304,000 Median High School Diploma Lifetime Earnings in 2009 dollars x 1.17 Inflation Adjustment to 2018 dollars per [a]) = 164,517
- ThrowawayR2 6y ago
  > "If there is one lesson from the last millennium, it's that innovation has poor economies of scale and is stifled by centralisation." When did we learn that? Bell Labs, Xerox PARC, and Intel's decades of dominance in semiconductor process research are prime counterexamples but I'm sure there are plenty of others.
  - JumpCrisscross 6y ago
    > Bell Labs, Xerox PARC, and Intel's decades of dominance Compare the size of these companies, in their time, to the Soviet-scale bureaucracies of Facebook, Apple, Amazon or Google. Innovation doesn't have no economies of scale. But they peter out around the levels of CERN or NASA. (At private enterprises, that threshold appears smaller.) Also note the difference between invention and innovation, the latter implying successful distribution. Xero PARC developing great things and then abandoning them isn't productive apart from fluke exfiltrations like by way of a tour.
wmf 6y ago
It might not be a bad thing if R&D becomes more decentralized back into startups instead of FAANGs. The money is there, but the current market structure doesn't encourage it.
klyrs 6y ago
The only "moonshot" I've seen from google is their quantum computing efforts. They held an internal competition, and axed one of the programs because it didn't look immediately viable. When management got bored, they rushed a "supremacy" result out the door (imo supremacy only counts when you're solving an existing problem, not when you're creating a problem that others can't solve, whose solution has no value). Their funding will be a question next year, too If Google's QC efforts fail, we've still got IBM, Rigetti, DWave, now Honeywell, and a smorgasbord of smaller efforts, clamoring for success. So if QC meets the fate of most other Google efforts, or gets lost in a breakup, will we lose much as a society? What other "moonshot" research do we see from FAANG? I can't really think of any, besides the insidious efforts to spy deeper into our homes and personal lives.
andrewjl 6y ago
Why should consumers subsidize R&D by a company with no accountability to them what so ever? US is the leader in many tech areas due to free market competition as well as defense spending, not Google's R&D. Google is also happy to sell their R&D to China, but then somehow finds it uncomfortable to transact with the DoD. Hard to make case Google is a net positive for the US with a straight face.
- actuator 6y ago
  They shouldn't, the accountability should come with regulation ensuring fair competition and respect for user data, not just going ahead and reducing competitiveness of US companies by breaking them up. I will be really surprised if Google sold their R&D to China, as it makes no sense whatsoever, but my comment was in general about all US big tech companies.
  - MiroF 6y ago
    > the accountability should come with regulation ensuring fair competition and respect for user data Yes, so what we're saying is there's clear evidence that they have failed at adhering to those regulation, so should be broken up. > not just going ahead and reducing competitiveness of US companies by breaking them up. Siri, define competition.
    - actuator 6y ago
      > there's clear evidence that they have failed at adhering to those regulation, so should be broken up. If there is clear evidence of them breaking existing regulation, then I am all for taking action. If the existing regulations are insufficient, I am all for introducing new ones as in the case of App Store debate. > Siri, define competition. I meant competitiveness in terms of R&D by cutting their cash flows.
      - MiroF 6y ago
        > If there is clear evidence of them breaking existing regulation, then I am all for taking action. Cool, so that is why they are taking Alphabet to court, because of alleged anti-trust violations. If the court determines there is clear evidence of them breaking existing regulation, then Alphabet might be broken up. Glad we agree that we ought to enforce our anti-trust laws.
ecf 6y ago
While America wants to break up their big tech, China is directly subsidizing their companies and shielding them from outside competition until they’re ready to launch on a global scale.
- sjg007 6y ago
  I think the two markets will stay disconnected somewhat. Maybe at the interface level -> e.g. Amazon/Walmart/Target
bleepblorp 6y ago
Most of the US tech sector doesn't spend all that much on R&D in comparison to its income and profits. For example, Alphabet reportedly spent $18 billion USD on R&D in 2018[0] but acquired $77 billion in gross revenues and $31 billion in net profits over the same year[1]. In comparison, Volkswagen spent $14.5 billion USD on R&D in 2018[0] but acquired only $54 billion (as converted from Euros at December 2018 exchange rates) in gross revenues and $13 billion in net profits[2]. If VW can spend that much on R&D, particularly as a percentage of revenues, without being an abusive monopoly, then so can Alphabet. [0] https://spendmenot.com/blog/top-rd-spenders/ https://spendmenot.com/blog/top-rd-spenders/ [1] https://www.marketwatch.com/investing/stock/googl/financials https://www.marketwatch.com/investing/stock/googl/financials [2] https://www.marketwatch.com/investing/stock/vow/financials https://www.marketwatch.com/investing/stock/vow/financials
- tomweingarten 6y ago
  This is meant to be satire, right? The percentage of revenues of the two examples you gave are nearly identical.
  - granzymes 6y ago
    23% vs 26%.... I hope it's satire!
- forrestthewoods 6y ago
  > Most of the US tech sector doesn't spend all that much on R&D 25% of revenue on R&D seems like a pretty reasonable amount?
- Nasrudith 6y ago
  Looking at R&D by inputs alone is a mistake anyway. You could spend billions on attemptint to revive phrenology but that doesn't mean you'll get anything of worth. There are certainly times when it needs more but spending more into it isn't an intrinsic merit. Plus even if you have a fruitful program it doesn't mean it can always be scaled up meaningfully just because there is more money available. The output of R&D is what justifies it - although great potential may take a lot of time to vindicate. A low ratio would result from a successful matured program which simply cannot expand fast enough to reinvest. Under that metric a program with the top people paid industry leading salaries and pumping out generations of advances would be worse than a lab full of scammers which never produced any applications. That isn't a very useful metric.

istorical 6y ago

creating new regulations on tech firms that make it more difficult for them to stop new competitors from arising or from protecting their monopolies makes so much more sense than just artificially deciding to slice up our top performers in industry and hamstring the sector.

RandomTisk 6y ago
That's assuming the monolith is actually doing good for society. I for one am desperate for more options for email providers with actual spam filtration, phones without anti-consumer spyware and eco system lockdowns, and social media that doesn't force users to speak only of things the DNC would approve of.
- Nasrudith 6y ago
  Haven't we learned from the death of Socrates that "good/bad for society" is a terrible and meaningless metric as it is so values dependent? All of those things already exist by the way - Google ironically can help you find it but so can most other search engines. There are plenty of paid providers in addition to roll your own and many anti-spam solutions. Fairphone/Freephone or even just some unlocked down Androids will run whatever you want and can, federated social media like Mastodon has all sorts of stuff that wouldn't be approved of by the mainstream like niche pornography.
- SpicyLemonZest 6y ago
  You're glossing over all the good things they've done to lay the groundwork for what you want. I'd like a smartphone with better privacy too, but I wouldn't want a smartphone at all if not for tech industry innovation over the past decade. I would (and did) choose a compact flip phone in a world where Blackberry was state of the art.

JumpCrisscross 6y ago

> R&D is costly, and these companies spend a lot on it To what end? And how efficiently? If there is one lesson from the last millennium, it's that innovation has poor economies of scale and is stifled by centralisation. Capital markets have not been more liquid in our lifetimes. In 2018 alone, venture capital "invested $251 billion," [1] representing the median lifetime earnings of almost 100,000 college graduates [2][3]. "25 percent of all VC deal volume" was in $1 billion+ deals, the median lifetime earnings of over 160,000 high school graduates [4]. Google is no SpaceX or Tesla or Moderna Therapeutics. It has its vanity projects. But those are run in start-up analogs. Google brings no advantage beyond its checkbook, and has a track record of its projects outperforming as spin-outs its internal operations. [1] https://www.mckinsey.com/industries/private-equity-and-principal-investors/our-insights/mckinseys-private-markets-annual-review https://www.mckinsey.com/industries/private-equity-and-princ... Venture capital’s very good year [2] https://cew.georgetown.edu/wp-content/uploads/2014/11/collegepayoff-complete.pdf https://cew.georgetown.edu/wp-content/uploads/2014/11/colleg... Figure 1, page 3 [3] $251 billion raised in 2018 / ($2,268,000 Median Bachelor's Degree Lifetime Earnings in 2009 dollars x 1.17 Inflation Adjustment to 2018 dollars per [a]) = 94,590 [a] https://www.usinflationcalculator.com https://www.usinflationcalculator.com [4] $251 billion raised in 2018 / ($1,304,000 Median High School Diploma Lifetime Earnings in 2009 dollars x 1.17 Inflation Adjustment to 2018 dollars per [a]) = 164,517

ThrowawayR2 6y ago
> "If there is one lesson from the last millennium, it's that innovation has poor economies of scale and is stifled by centralisation." When did we learn that? Bell Labs, Xerox PARC, and Intel's decades of dominance in semiconductor process research are prime counterexamples but I'm sure there are plenty of others.
- JumpCrisscross 6y ago
  > Bell Labs, Xerox PARC, and Intel's decades of dominance Compare the size of these companies, in their time, to the Soviet-scale bureaucracies of Facebook, Apple, Amazon or Google. Innovation doesn't have no economies of scale. But they peter out around the levels of CERN or NASA. (At private enterprises, that threshold appears smaller.) Also note the difference between invention and innovation, the latter implying successful distribution. Xero PARC developing great things and then abandoning them isn't productive apart from fluke exfiltrations like by way of a tour.

wmf 6y ago

It might not be a bad thing if R&D becomes more decentralized back into startups instead of FAANGs. The money is there, but the current market structure doesn't encourage it.

klyrs 6y ago

The only "moonshot" I've seen from google is their quantum computing efforts. They held an internal competition, and axed one of the programs because it didn't look immediately viable. When management got bored, they rushed a "supremacy" result out the door (imo supremacy only counts when you're solving an existing problem, not when you're creating a problem that others can't solve, whose solution has no value). Their funding will be a question next year, too If Google's QC efforts fail, we've still got IBM, Rigetti, DWave, now Honeywell, and a smorgasbord of smaller efforts, clamoring for success. So if QC meets the fate of most other Google efforts, or gets lost in a breakup, will we lose much as a society? What other "moonshot" research do we see from FAANG? I can't really think of any, besides the insidious efforts to spy deeper into our homes and personal lives.

andrewjl 6y ago

Why should consumers subsidize R&D by a company with no accountability to them what so ever? US is the leader in many tech areas due to free market competition as well as defense spending, not Google's R&D. Google is also happy to sell their R&D to China, but then somehow finds it uncomfortable to transact with the DoD. Hard to make case Google is a net positive for the US with a straight face.

actuator 6y ago
They shouldn't, the accountability should come with regulation ensuring fair competition and respect for user data, not just going ahead and reducing competitiveness of US companies by breaking them up. I will be really surprised if Google sold their R&D to China, as it makes no sense whatsoever, but my comment was in general about all US big tech companies.
- MiroF 6y ago
  > the accountability should come with regulation ensuring fair competition and respect for user data Yes, so what we're saying is there's clear evidence that they have failed at adhering to those regulation, so should be broken up. > not just going ahead and reducing competitiveness of US companies by breaking them up. Siri, define competition.
  - actuator 6y ago
    > there's clear evidence that they have failed at adhering to those regulation, so should be broken up. If there is clear evidence of them breaking existing regulation, then I am all for taking action. If the existing regulations are insufficient, I am all for introducing new ones as in the case of App Store debate. > Siri, define competition. I meant competitiveness in terms of R&D by cutting their cash flows.
    - MiroF 6y ago
      > If there is clear evidence of them breaking existing regulation, then I am all for taking action. Cool, so that is why they are taking Alphabet to court, because of alleged anti-trust violations. If the court determines there is clear evidence of them breaking existing regulation, then Alphabet might be broken up. Glad we agree that we ought to enforce our anti-trust laws.

ecf 6y ago

While America wants to break up their big tech, China is directly subsidizing their companies and shielding them from outside competition until they’re ready to launch on a global scale.

sjg007 6y ago
I think the two markets will stay disconnected somewhat. Maybe at the interface level -> e.g. Amazon/Walmart/Target

bleepblorp 6y ago

Most of the US tech sector doesn't spend all that much on R&D in comparison to its income and profits. For example, Alphabet reportedly spent $18 billion USD on R&D in 2018[0] but acquired $77 billion in gross revenues and $31 billion in net profits over the same year[1]. In comparison, Volkswagen spent $14.5 billion USD on R&D in 2018[0] but acquired only $54 billion (as converted from Euros at December 2018 exchange rates) in gross revenues and $13 billion in net profits[2]. If VW can spend that much on R&D, particularly as a percentage of revenues, without being an abusive monopoly, then so can Alphabet. [0] https://spendmenot.com/blog/top-rd-spenders/ https://spendmenot.com/blog/top-rd-spenders/ [1] https://www.marketwatch.com/investing/stock/googl/financials https://www.marketwatch.com/investing/stock/googl/financials [2] https://www.marketwatch.com/investing/stock/vow/financials https://www.marketwatch.com/investing/stock/vow/financials

tomweingarten 6y ago
This is meant to be satire, right? The percentage of revenues of the two examples you gave are nearly identical.
- granzymes 6y ago
  23% vs 26%.... I hope it's satire!
forrestthewoods 6y ago
> Most of the US tech sector doesn't spend all that much on R&D 25% of revenue on R&D seems like a pretty reasonable amount?
Nasrudith 6y ago
Looking at R&D by inputs alone is a mistake anyway. You could spend billions on attemptint to revive phrenology but that doesn't mean you'll get anything of worth. There are certainly times when it needs more but spending more into it isn't an intrinsic merit. Plus even if you have a fruitful program it doesn't mean it can always be scaled up meaningfully just because there is more money available. The output of R&D is what justifies it - although great potential may take a lot of time to vindicate. A low ratio would result from a successful matured program which simply cannot expand fast enough to reinvest. Under that metric a program with the top people paid industry leading salaries and pumping out generations of advances would be worse than a lab full of scammers which never produced any applications. That isn't a very useful metric.