Google Canceled the Launch of a Robotic Arm After it Failed the 'Toothbrush Test' (bloomberg.com) 97
Mark Bergen, reporting for Bloomberg: Google published research this week detailing how its software enables robots to learn from one another. To demonstrate, the company's scientists showed videos featuring robotic arms whirling inside its labs. Google's robotics group built those machines and wanted to sell them to manufacturers, warehouse operators and others. However, executives at Google parent Alphabet Inc. nixed the plan because it failed Chief Executive Officer Larry Page's "toothbrush test," a requirement that the company only ship products used daily by billions of people, according to people familiar with the situation.
Pretty short sighted (Score:2)
Re:Pretty short sighted (Score:5, Informative)
According to the article, the research isn't over. They just aren't going to sell the robotic arm. This doesn't even mean the robotic arm won't get sold--if Google decides it has no use for this, it can just sell the IP to someone else to develop and bring to market. Google just doesn't want to be in the robotic arm selling business right now.
Re: (Score:2)
Comment removed (Score:4, Interesting)
Re: (Score:2)
Re: (Score:1)
Because they're trying to unfuck our spelling by making it phonetic.
And it's could've if you're writing conversationally (IMO).
Re: (Score:2)
How is the proper pronunciation of "have" so unphonetic that it is better written as "of". Have starts with an "H" sound as in ha-ha or he. Then the "A" sounds just like in "at", "hat", "cat", etc. then we have the nice little "V" sound. While true that common American pronunciation of "could have" becomes more like "uv" with the "U" sounding like "ugh" or "up". "Could of" comes nowhere near that.
Re: (Score:1)
Because nobody would say "have" in that situation, they'd say " 've"
the word "of" has it's own issues, but one at a time.
Re: (Score:2)
Making it phonetic would actually make it closer to "could af", where does the "o" come from?
Re: (Score:1)
I think "could've" is probably best, both phonetically and conveying what's said, but could of is certainly closer than could av is closer than could af.
But I guess part of the problem with phonetic spelling is that it's regional...
Re: (Score:2)
Except people rarely say could have, usually they say could've, and the d-v transition mimics the sound of an intervening o or u. And for whatever weird historically reason, of is pronounced uv.
Re: (Score:2)
You might want to get out more. Not everyone lives on the same block you do, and some people are actually literate. Most people do say "could have" round here.
In the olden days, I ran a BBS (like a blog, but old tech). I had people who spoke no English complaining of spelling mistakes or typos because they literally looked up every single word in the dictionary. I got the message.
These days, they probably use Google translate, but that is pr
Re: (Score:2)
An interesting mix then - I assume they say could not instead of couldn't as well? And yet I would assume they still manage to understand your usage of round (or, more properly, 'round) rather than around.
Yes, the specifics of dialect vary wildly within the language, but pretty much everyone understands and uses the standard contraction forms. For could have, that's could've, a homophone of the ungrammatical could of, rather than could'ave, which is just wholly wrong in written English (though I have no d
Re: (Score:3)
I assume they say could not instead of
For the last time, it's "could not instead've!"
Re: (Score:2)
Some of us also learned English. And that entails learning proper pronunciation.
I wouldn't go as far as pretending I know better when to use what past tense than a native speaker, but with some people I would actually dare to...
Re: (Score:2)
Sure if you like to waste time. In Northeastern US, people who expand contractions in daily speech are either being pretentious, or hicks.
They are doing it wrong (Score:4, Funny)
It did not fail the test. It does not want to or need to brush its teeth.
Re: (Score:3)
Why start the project then? (Score:2, Insightful)
If this was a known requirement for googles products why did they start to begin with
Re:Why start the project then? (Score:5, Interesting)
Google is a little like 3M in that it's an internal project machine. They fostered a culture of internal innovation that wasn't dependent on deliverable products until a certain critical size is reached. When it gets there it gets evaluated to see if there's value in continuing it.
Sometimes the answer is yes, sometimes the answer is no.
I remind you that 3M rejected the idea of the Post-It and didn't fund it internally either, but they did fund the laser disk.
Re:Why start the project then? (Score:5, Informative)
If this was a known requirement for googles products why did they start to begin with
There are applications for robots that could potentially be sold to billions of people. This just isn't one of them. A household cooking & cleaning robot would be a good product for Google. Warehouse automation robots are not.
Re: (Score:3)
Here, I'll read the article for you. The "toothbrush test" = requirement that the company only ship products used daily by billions of people.
Oh, what's that you say? I didn't even have to read the article; it's right there in the summary? Next time maybe they should just put the whole summary in the title.
Re: (Score:2)
It says right there in the summary: "a requirement that the company only ship products used daily by billions of people." Billions of people don't use robotic arms every day, nor are they likely to start anytime soon.
On the bright side, at least the robot didn't have to pass the breakfast test [youtube.com].
Re: (Score:1)
if he isn't even going to read the summary we should just make something more interesting up.
such as...
it means that the robot arm couldn't be taught to brush it's own teeth so they decided they can't sell it because they didn't think the customer would properly care for the robots teeth.
Re: (Score:2)
What the fuck is a toothbrush test?
Toothbrushes are needed by everybody alive and they wear out frequently so you have to buy multiple toothbrushes over time.
The toothbrush test is basically about how many of the developed item can/will you be able to sell?
Google is S00000000 smart... (Score:2)
that they rolled the meter back to 00000000 and now they are dumb again.
Re: (Score:2)
dumb again
I wish I would be as dumb as Google. I wouldn't even need to laugh to the bank. I could just pay the bank to come to me, and pay someone else to laugh at it when it gets here.
One Size Fits All (Score:5, Interesting)
Clearly if it has the word Google on it, it must be WEBSCALE!!!
Really, have an off-brand trade name for products that are niche industries if you've got the stink on for smaller products. Seriously, Google's scorched earth approach for lower performing products has affected my love for the company significantly over the last couple years. I'm VERY leery to try any of their new offerings, which is clearly a self-fulfilling prophesy.
Re: (Score:2)
If you could sell me a cheap robotic arm that could learn to do some simple* tasks, I'd probably get some for washing dishes and folding laundry. With a
Google Glass (Score:1)
So did they think Google Glass would be used by billions of people? LOL!
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Disappointed (Score:4, Insightful)
I am disappointed. From the headline I was expecting that they used the robotic arm to toothbrush someone and then things got funny...
Re: (Score:2)
Clearly everyone needs this arm to brush their teeth for them. Add in shaving and combing and flying cars and we can live like the Jetsons!
Re: (Score:1)
Pffffffft! A robot wouldn't user razors, it would use lasers!
If Alphabet doesn't want to do it, sell it off! (Score:5, Interesting)
Why wouldn't Alphabet spin off a new company that they have a 40% stake in and let it fly?
It wouldn't be part of Alphabet, so the rules wouldn't apply.
If it fails, they can handle a little loss.
If it is a hit, they can make money from it without holding back on good ideas the world might be able to use.
Re: (Score:2)
Exactly. They wouldn't even need to retain 40%, just sell it completely.
There'd be IP and potentially licensing complications, but a few million (or tens or hundreds of millions) in cash or stock from a purchaser covers a lot of inconvenience.
TFA suggests they're looking to spin off Boston Dynamics but that doesn't excuse cancelling the programme. The alleged justification around brand is also nonsense, it's trivial to create a new company and brand, even 100% owned.
But hey, I don't run a multi-billion doll
Re: (Score:3)
Why wouldn't Alphabet spin off a new company that they have a 40% stake in and let it fly?
It wouldn't be part of Alphabet, so the rules wouldn't apply.
If it fails, they can handle a little loss.
If it is a hit, they can make money from it without holding back on good ideas the world might be able to use.
That is why I clicked on this thread. There should be room in the Alphabet ecosystem to spin off R&D like this into a standalone business or to sell to other companies for further development if there is a viable business plan and a sizable enough market.
Otherwise they will get stuck in the mindset that befell Xerox's Palo Alto Research Center when it was at the forefront of computing R&D and ended up with other companies actually developing their concepts because the concepts for computing didn't
Re: (Score:2)
until i read the summary i envisioned it fumbling to pick up a toothbrush that it dropped endlessly
The toothbrush test is idiotic (Score:5, Interesting)
There are huge industries built around things that fail the toothbrush test. Does Google not like getting a return on their investments?
Re:The toothbrush test is idiotic (Score:5, Interesting)
Google's entire business model is built on massive scale. Small-but-hugely-profitable industries require a completely different approach to dealing with clients and users. If a user is having trouble with, say, their Nest thermostat, Google can log the problem and work on a fix in their own time. If an industrial client is having trouble with a robot arm, Google would be expected to be dedicating resources to that specific arm within twenty-four hours, if not sooner.
Re: (Score:3)
I think Google could absolutely build industrial robotics that's at least as good as what's currently out there. I don't think Google wants to build the support system that such products would require. It's a massive commitment that would demand a pivot from how Google typically deals with customers and manufactures products. Nest and Google Home is not even close.
It's not that they couldn't- especially since the structure of Alphabet is perfect for it- but the company seems very hesitant to branch out i
Re: (Score:2)
They can always sell the design to a different company with expertise in this type of market.
Also, I initially imagined the "toothbrush test" as a requirement that robotic arms be able to brush someone's teeth autonomously.
Re: (Score:2)
That's entirely correct. There are other implications when selling machinery, i.e. capital goods: the individual client has much more power over the manufacturer. A toothbrush consumer represents only one-billionth of your revenue and has virtually no power over the manufacturer. A capital goods customer can represent several percent of your revenue - in some industries several ten percent. That is on the order of the operating profit, i.e one customer can influence a lot the economic outcome.
To deal with s
Pixel (Score:1)
Re: (Score:2)
Google would probably argue the Android OS is the "product" in this case, and since it is already on billions of phones used every day it meets the test.
Thought they were going a different way with that (Score:3)
When I saw the phrase "robot arm" in conjunction with "toothbrush test", I had horrible visions of mangled faces from the various trials where the powerful robot arms went out of control during delicate teeth-cleaning operations were in progress.... *shudder*
Re: (Score:2)
I envisioned the exact same scenario. I hoped no-one was tempted to ask the robot for a handjob.
Re: (Score:2)
Hooli beat them to it (Score:2)
This scene with the monkey arm in Silicon Valley is uncannily precedent. I can't believe no one has mentioned this yet:
https://youtu.be/1KaWPYOLuT8 [youtu.be]
Re: (Score:2)
Re: (Score:2)
It was prescient to know that his mistake would be pointed out, this being Slashdot
The toothbrush test? (Score:3)
used daily by billions of people
Uh ... I can think of something that billions of people "use" daily. And that, plus a robotic arm, equals Internet Rule 34.
'Scuse me, I'm going to brush my teeth.
Re: (Score:2)
Life imitating the THX 1138 director's cut? Now you've gone too far.
Alternative (Score:1)
I don't need a bot to brush my teeth, I have it done
by bigfoot. [slashdot.org]
So it's toothbrushes? (Score:2)
Not hotcakes like it was in my day?
Get off my lawn...
Even worse than that (Score:1)
It didn't floss
That's a relief... (Score:2)
For a moment there, I had a horrible premonition [youtu.be]
This kind of project is not for making products.. (Score:2)
But rather for creating parts you use in products.
Basically they have to squander all they got from it, do a few more of those, and then assemble all into something that can be sold.