Waymo wants Uber to pay $2.6B in damages, just for starters | Hacker News Reader

Waymo wants Uber to pay $2.6B in damages, just for starters

arstechnica.com

109 pointsnwrk45 comments

Can anyone comment on how Waymo might come up with such a figure? Especially in the context of something that hasn't materialized yet (no revenue)? I assume this is towards the high-end of what they expect to get. Also, why would Uber want to push for making this figure public? Seems like it would be against their interests at this point.

The figure is estimated based on the investement google has made.

Didn't we just read Google has invested $1.1B into self driving and a significant portion of that was paid to Levandowski?

It'd be based on a more comprehensive view of total damages. The investment already made, the present estimated value (they'd probably argue it's a multiple of the original investment), and some amount of future potential value that Waymo may be deprived of due to Uber's actions (not sure how well that would hold up here, it's a very unclear extrapolation, unlike say a business with consistent earnings over many years).

The amusing thing is that Levandowski comes out of this just fine. As a Google employee, he was required to sign an arbitration agreement. That cuts both ways - Google can't sue him. Google did take him to arbitration, and there was some settlement, but apparently not a huge one. That's over, and Google can't reopen that matter.

The arbitration didn't involve trade secret claims, which the lawsuit against Uber does. Instead, the arbitration involved poaching employees.[0] Poaching employees is child's play compared to stealing the LIDAR design.

Second, Judge Alsup referred the case the US attorney's office for criminal investigation. If convicted of intellectual property theft (and there's a lot of evidence about this), Anthony Levandowski could be end up in prison for a very long time.

[0] https://techcrunch.com/2017/03/29/waymo-pursued-arbitration-...

[1] https://www.wired.com/2017/05/googles-fight-uber-takes-turn-...

>"If convicted of intellectual property theft (and there's a lot of evidence about this), Anthony Levandowski could be end up in prison for a very long time."

That's just barbaric. Yes, IP laws serve a public good when they're able to balance the needs of incentivizing creative endeavors with ensuring those endeavors eventually end up in the public domain. But it's not even remotely morally defensible to put a human being in a fucking cage for years because they shared some secrets related to making self-driving cars.

The US already has way too non-violent people in prison.

The amusing thing is that Levandowski comes out of this just fine

I'm pretty sure he regrets what he has done. About half of his money has gone to taxes and the rest has a way of going pretty fast when you have a Google and DOJ on your back. So, I doubt that he will be fine.

He can surely manage... even with half of his 120 million dollar bonus going to taxes.

That's 60 million, assuming that anything prior to that is relatively pocket change. Lawyers cost $1000+ an hour and you need several, for years. Not to mention any possible claw-back, lawsuits for fraud, theft etc etc. There was an article yesterday here; Lance Armstrong has spent some $15 million on his lawyers alone so far.

Maybe they want to have the figure public because they see it as very high and want to bias the public about the absurdity (from their perspective) of such a high number? People may see the number and conclude that Waymo is just being vindictive/aggresive/jerks and believe the suit has no merit.

I have absolutely no insider knowledge and am just speculating.

The estimated figure will be calculated similar to how a startup's value is calculated. There is some assumption made based on current market size, say the total car sales, and potential market growth. They then work backwards to calculate the estimated value of such technology at present date.

As for making the figure public and it is pure speculation. It might be an attempt to reflect this on Waymo being greedy - as you might have noticed the by line "just for starters". Most people tend to read headlines and not content. Secondly it might give them an edge during jury selection for the trial.

What I'm also interested in is how much money do lawsuits like these cost. Both companies have large amounts of cash, so this isn't one of those cases where the one with the most money wins, but I do expect both of them to spend a lot.

Since we don't have full disclosure of what tread secrets uber is being accused of but it is a general knowledge that it revolves around Grizzle bear 3 circuit board design. There had been two recent stories which tried to portray that the whole thing was of low-value to Waymo.

https://www.bloomberg.com/news/articles/2017-09-06/alarm-bel...

and this ridiculous story of previous version of this circuit board was gifted to departing employee in the form of earrings:

https://spectrum.ieee.org/cars-that-think/transportation/sel...

Both of these stories points to waymo not being diligent in hiding their tread secret or simply did not value this information much. That's why Uber is probably trying to make this figure public to make them sound greedy.

Being pre-revenue doesn't mean something has no value. They are calculating it based on what value they deem it to have. If the stolen tech is of vital importance to producing a product that could be worth a nearly 3 billion dollars, then that's what they'll sue for. Actually they'll sue for a bit more, but that's roughly how it works. Like anything, it's better to ask for $10 when someone is only willing to pay $8, than to ask for $5 when someone is willing to pay $8.

While the theft of trade secrets central to developing self-driving cars obviously seems important, and this case might have a significant impact on the future market for autonomous vehicles, it's not clear to me where these damages numbers are coming from.

Neither Waymo nor Uber have actually made much money from self-driving cars at this point - Uber has its small-scale pilot programs, and Waymo has R&D partnerships with companies like Avis and Lyft. But unless there are some major potential deals that Waymo missed out on because of Uber's mere presence and their quick development of self-driving technology, I can't imagine they'd be able to argue that they've been damaged at this point - this isn't like a patent case where companies are making money off their products.

It'd be interesting to see a more unusual argument from Waymo, though - something like "The trade secrets Uber stole allowed them to more easily recruit engineers, reducing the talent pool available to us and delaying the wide release of our own product."

Maybe someone here knows the answer to this:

How much of this stolen stuff would already be disclosed (or soon to be disclosed) in patents?

Do they really just try and protect these inventions largely as trade secrets?

If it's largely covered by patents, then obviously there is some value in knowing what has been tried/has actually worked. But doesn't patent coverage give them Google/Waymo a stronger way to protect these inventions?

(please note I'm not arguing for/against software/algorithm patents, just trying to better understand)

Patents start from the filing date and last twenty years. Some of these things Google has could be several years old already, and it's unlikely that self driving cars will be on the road for at least another few years. Patenting something in 2010 that won't be profitable until 2020 means you've killed half the lifetime of the patent, so in that case a trade secret may be much better.

Broad strokes (and very simplistic) philosophy...

Patents are most useful for inventions that can be easily reverse engineered. If the competition can disassemble a unit and create their own derivative from the knowledge gained, a patent blocks the use of a derivative while not revealing anything that would not be revealed by reverse engineering anyway.

If an invention is very difficult to reverse engineer, it is better to keep it a trade secret because the competition (ideally) will not be able to create a derivative via reverse engineering. In this case, a patent will reveal information that a competitor would otherwise not be able to discover e.g. by reverse engineering.

Isn't the worry then that someone else comes up with the invention independently and patents it? (and does the US being a first-to-file patent system have an impact on this)

The patent would probably be invalid if the product was already on the market from a different vendor.

Hasn't (or isn't) the USPTO switching to a first-to-file system? Proving prior art using a trade-secret is likely to be a very challenging proposition (due to the 'secret' bit). IANAL.

Right, but if the product is on the market, doesn't that make the whole trade secret thing kind of void anyway? Because it most likely can be reverse engineered?

If it's provided to users only as a service, then I can see they could protect it as a trade secret by limiting access somehow. But then... would that invalidate the patent (considering the first-to-file stuff).

The classic example of where a trade secret works well is for the recipe (original) for coca-cola.

Well that and the fact that nobody else is allowed to import one of the known ingredients, coca leaves: https://en.wikipedia.org/wiki/Coca-Cola#Use_of_stimulants_in...

Interestingly, isn't that the whole runway that they've got left?

Uber has 5.5B in cash, and 3B in credit available

I would guess the 3B in credit is no longer available if they have 0 cash left

I would guess they'll draw on the 3B in credit well before then.

Finally, a viable business model for self driving cars

I laughed, but I'd prefer to keep the low-effort comedy posts to Reddit if possible.

I'd prefer to keep the low-effort passive-aggressive insult posts to Reddit if possible.

I don't understand HN's contempt for Reddit. Reddit has lots of good stuff. If all you see in Reddit is the low-effort posts, maybe you're the one spending too much time in low-effort subreddits.

Is it just me or does this sound like the typical big company coming out of the woodwork to sue to sabotage someone else's IPO. Isn't it just jump the shark moment for Google?

I find HN sentiment too favorable to Google these days, compared to Yahoo's lawsuit against FB at their IPO time. I think this theory should at least be considered.

Wait, Uber aided and abetted theft. IP theft, but not software, this is hardware, classic, very little disagreement, IP theft. Why shouldn't Google ask society to impose a high cost on such behavior. Seems to me, if anyone believes anything about the patent system in it's classic form makes sense, then they should support a high cost to IP theft.

IP... Imaginary Property.

"Theft" is not a technical jargon. It's an emotionally charged phrase to trigger public opinion in favor of patent monopolists.

Also, didn't the judge throw out the patent case and the dispute is merely trade secrets now?

If what I create with my mind isn't my property, then what you create with your hand isn't yours either.

But you knew that already. You're just trolling, I'm sure.

An alternative point of view is that no deep a priori justification for the concept of ownership exists, other than that you own what you can actually defend as such, be it by force or by law. Under this point of view, it would be totally conceivable that you couldn't own the products of your mind but could those of your hands, if the law just said so.

And the reverse as well. Perhaps intellectual property is the only kind that matters.

> If what I create with my mind isn't my property, then what you create with your hand isn't yours either.

That implication is not a logical necessity. There is a big distinction: if I take away the thing you made with your hands from you, you have one fewer thing, and I have one more thing.

If I learn or copy what you have in your mind, both of us have something. I cannot take it away from you.

Also, the whole phrase "Intellectual Property" is a misleading thing that should not be used in any reasonable discussion, because it derails the arguments by unifying a bunch of pretty different things[1].

We should be calling copyright, patents, and trademarks what they are: they are not properties, they are monopolies authorized by the government (trade secrets may be a bit distinct here). Perhaps authorized monopoly is a more accurate term than "intellectual property". cf. [2]

[1] https://www.gnu.org/philosophy/not-ipr.en.html

[2] http://www.locusmag.com/Perspectives/2016/11/cory-doctorow-s...

Ok, but to the original point, which was Uber aiding and abetting Levandowski's taking away specific documents (do you admit that a document is a thing?) which were trade secrets: surely that qualifies as theft of valid intellectual property.

Also, you, like almost everyone in this economy, makes their money on the work product of their mind. Some is more pilferable than others, but if you disagree, how about you just share your screen with me at all times, and I can share that screen with whomever I want?

"svn checkout" of something you were a core developer of and could fetch 14k files.

Not negating your point but putting it in perspective. Downloading and archiving your own work email inbox is also far from uncommon I'm sure. Doesn't imply you definitely have a specific use for it.

Far from uncommon among folks preparing to steal company property when they quit. I don't know anyone who downloads their email archive when the situation is normal.

Regardless of the accuracy of the phrase, it seems to be effective at communicating what I mean.

And as we all know, what you'd be taking away from me isn't knowledge but profit from that knowledge. I'm now reminded of those fairy tales where the witch takes the promise of someone's future child.

This assuming Uber's IPO is around the corner? While a case could be made against Yahoo during Facebook's IPO (Yahoo sued FB on March 12, 2012 and the IPO was filed on February 1, 2012), situation here is not similar.

From what I read Uber has discussed about IPO prospects but 1.5-3 years later: https://techcrunch.com/2017/08/30/new-uber-ceo-says-ipo-is-1...

So what effect does it have really on Uber's IPO?

Reply on news.ycombinator.com