Google Bid Pi Billion Dollars For Nortel Patents 213
mikejuk writes "Google mystified other participants in an auction for patents last week by their choice of bids. They weren't the round regular numbers that are normally expected. After first bidding $1,902,160,540 — a reference to Brun's constant — and later bidding $2,614,972,128 for the Meissel-Mertens constant, they ended up submitting a bid for $3.14159 billion. Google ended up losing the auction — but was that a deliberate ploy?"
The deal fails when Nortel askes for exact change. (Score:5, Funny)
The deal fails when Nortel askes for exact change.
Re: (Score:2)
CFO's glad they didn't take the next step (Score:4, Interesting)
Google's CFO's glad they didn't take the next step after pi: tau [google.com] (6.28...)
Re:CFO's glad they didn't take the next step (Score:4, Funny)
Google's CFO's glad they didn't take the next step after pi: tau [google.com] (6.28...)
The CFO's would have been more worried at a bid for $googol [google.com].
Re: (Score:2)
They could have just gone after 4.66920 the Feigenbaum constant
Re: (Score:2)
You're clearly misinformed. Tau is a much more elegant number for every use case. Way to post AC, btw.
Re: (Score:2)
You're clearly misinformed. Tau is a much more elegant number for every use case. Way to post AC, btw.
So you're claiming that it is better for calculating areas of circles? (\pi r^2 vs \tau/2 r^2) Or volumes of spheres? (4/3\pi r^3 vs 2/3\tau r^3)
Re: (Score:2)
The volume of a sphere is actually 1/6 tau r.
Re: (Score:2)
Re: (Score:3)
My problem was that the Slashcode sucked up the ^3 character.
Re: (Score:2)
Want to take a shot at enlightening the rest of us non-tau plebes?
Re: (Score:2)
Just read this, and you'll at least note how many mathematical operations are simplified with a constant that means "a whole turn through a circle": http://tauday.com/ [tauday.com] Tau = 2*Pi, just a single constant value off, so the math isn't much more difficult to do, but the meaning of what you're doing can be masked by that constant.
Tau as a symbol is already used in math related to pi (pi/2, not 2pi). We need to start using futhark runes or something.
Re: (Score:3)
Calling people "halfwit" (or any other name) is the hallmark of someone who doesn't know what they're talking about.
You're confusing the idea that an expression has fewer terms with the idea that an expression is more elegant. That is simply not true. The core of elegance is conceptual clarity and simplicity.
A circle is a curve. How do you find the area under a curve? Oh wait... and what do integrals look like for these kinds of expressions and relationships? Let's walk through this conceptually: the area p
Re: (Score:2)
You could at least tell us non-math-degrees...
Re: (Score:2)
And everyone knows that red is the best colour, because red wunz go fasta [wikipedia.org].
Re: (Score:2)
Tau seems more mathematically correct given the nature of calculus and the relationship between instantaneous slope and areas under curves, or derivatives and integrals, respectively.
Consider the equations of motion of an object given a constance acceleration, C.
Accel = C
We know the relative velocity through integration with respect to time.
Velocity = C * t
We know the relative position through integration with respect to time.
Position = 1/2 C * t^2
As we move from integrating and differentiating with respect
Re: (Score:2)
Tau makes doing math much MUCH easier. Quick, what angle (in radians) corresponds to 1/128 of a circle?
(Hint: It's (tau)/128)
Re: (Score:2)
Tau is not "wrong". It accurately describes exactly what it intends to describe.
Everything else is just usage preference. And frankly, that's a matter of personal taste. There's nothing wrong with saying "Tau" when you might otherwise have written "2Pi", although it isn't intrinsically more right either.
If you use maths for a living, you should know that written form maths is a flexible thing; there are usually a number of ways of going about the same thing, and one way is usually only nominated as "right"
Weird bid numbers are normal for large bids (Score:5, Interesting)
If you're willing to bid $3 billion for something, the last thing you want is for someone to bid $3 000 000 001.00, and beat you in the bid. So it is quite routine for bids on large contracts with a closed bidding process to use unusual numbers rather than round numbers. I've seen this in multi-tens or hundreds of million-dollar land acreage bids for the rights to drill for oil. They'll bid $30 545 777.88, and weird things like that. Usually the "extra bit" is a small percentage of the total bid amount, but if you're going to do that, why not have some fun with it? And if you're "mystifying" the other participants, good! That's the whole point -- to keep them guessing and prevent them from figuring out your strategy so they can't bid $1 more.
Re: (Score:2)
It's common when bidding on Ebay too. If your opponent "thinks round", act differently.
Re: (Score:2)
It's common when bidding on Ebay too. If your opponent "thinks round", act differently.
It's a bit harder to get creative with the few digits you usually have on eBay :)
Re: (Score:2)
I've done it on eBay. It was a good strategy there against bid sniping. eBay prioritises early bids, so if you specify a limit of £20 and someone else does the same later then you win at £20. If someone jumps in at the last second with £21, then they win and you wouldn't see their bid before the auction ended[1]. Quite often, they'd bit £20.01, and then they'd get it for 1p more than you said you were willing to pay. If you bid £20.15 or something, then the bid would go up
Re: (Score:2)
Re: (Score:3)
No. the entire idea behind ebay is to bid the maximum amount you are willing to pay. If you do that there is NEVER a reason to make a 2nd bid.
OK, but you're competing with people who often don't follow this rule.
In general, you have two goals -- 1) win the item, and 2) pay as little as possible, and in general, the best way to achieve these goals is to bid exactly once -- at the last possible moment -- with the highest amount you're willing to spend on the item.
You could also bid this same amount earlier in the auction, but that bid by itself changes things -- simply by bidding you're telling other bidders that this item is valuable, and they may be willing to up their bids based on this new information. You do best by denying them this additional information until it's too late.
(Yes, if there's two equal bids, the earliest one wins -- but that's easily defeated by bidding an odd amount. Don't bid $100.00 -- bid $104.26, which will beat any bid up to $104.25, no matter when it was placed.)
Re: (Score:2)
the entire idea behind ebay is to bid the maximum amount you are willing to pay.
And the entire idea behind auctions is that people's willingness to pay depends on what other's are willing to pay. Thus eBay is a self-contradiction, and that's why it is so frustrating.
Re: (Score:2)
The thing is that what you say would be correct in the ideal world where everybody understood E-bay. In real life, the other people do bid twice; they don't undersand how E-bay works and it is worth your while to do things a bit differently. If they see that you outbid them, the rethink and often bid higher. This pushes up the price you have to pay. These people have bid incorrectly; it's true; maybe if they understood ebay they would have bid higher, but it's not in your interest to let them have the
Re: (Score:2)
The thing is that what you say would be correct in the ideal world where everybody understood E-bay.
It's not that people don't understand e-bay, it's that people aren't rational and what they are willing to pay changes depending on the current valuation of the item. If they see something on ebay that's currently going for $1.50, they might say, "I'm willing to pay up to $10 on that." Once they get outbid and the price climbs to $11, they become emotionally invested and adjust their maximum. They start thinking, "well, maybe I can go up to $20," even though they would have never agreed to go up that hig
Re: (Score:2)
Re: (Score:2)
I do think that if the people understood E-bay, they would understand that their original bid should be their entire valuation and then they would not make the second bid.
Well, you're absolutely right. I was just trying to point out that they also need to understand the emotional component that makes them want to bid irrationally. I think it's the same thing that happens in poker. You put a large amount in the pot, and you feel like you're invested in that hand. if you're a smart player and you notice the odds are against you making that hand, you know that the money in the pot is already gone. Beginners will completely ignore pot odds once they're emotionally attached
Re: (Score:2)
Isn't that the way all auctions work, not just eBay?
Re: (Score:2)
"Even more interesting is that the existence of snipers is probably good because it means that early bidders come to understand quicker the way e-bay auction system and the need to bid their maximum bid on their first bid."
That's illogical. If that were true, the practice of sniping would gradually decline as people learned--but that's not the case. Sniping works because the system rewards it. Until that changes, sniping won't go away.
What is needed is an option for sellers to use that would keep an auct
Re: (Score:2)
THe problem is, you give away your maximum bid informatio
Re: (Score:2)
citation needed
Sounds unwise (Score:2)
At best, they were engaged in an advertising campaign that had the potential for being extremely expensive (marketing cost = magic number that became their winning bid - lowest bid that would have won). At worst, they were being extremely foolish with shareholders' money, potentially overpaying by hundreds of millions of dollars just because they think some numbers are cool.
Google: Global Superpower Math Nerds (Score:3, Funny)
This is exactly the kind of behavior I would expect from a group of guys who, once routinely stuffed into their high school lockers, have now grown up (?) to become full-fledged white cat-stroking Bond villains.
I give it another 6-9 months more of federal government inquiries and subpoenas before they dig a moat around their campus and fill it with laser-headed sharks...
Re: (Score:2)
What, you mean their Google Apps datacenters don't count? Watching the tour video on youtube, and I think their getting close: Everything from fingerprint/eye readers to camera banks and 24/7 security guards.
Rounding (Score:2)
Well, normally I think companies try and figure a ballpark figure that might be good for their bid. They then round it arbitrarily to "round" numbers that are near their ballpark figure. This is susceptible to the same overpaying problem in that a round number may be greater than the lowest bid the would have won.
In this case, they were just using universal constants as their round numbers.
Re: (Score:2)
Well, normally I think companies try and figure a ballpark figure that might be good for their bid. They then round it arbitrarily to "round" numbers that are near their ballpark figure. This is susceptible to the same overpaying problem in that a round number may be greater than the lowest bid the would have won.
In this case, they were just using universal constants as their round numbers.
Fair point. If you're right, then MBA's preference for bid amounts ending in lots of zeros is no more efficient than Google's preference for numbers ending in digits other than zero.
Re: (Score:2)
Maybe they were playing a kind of brinksmanship, forcing competitors to bid higher and higher for patents they needed more than Google did, and thus depriving competitors of cash useful for other operations that are more critical to Google.
This kind of corporate strategy makes a lot of sense for shareholders -- if you win the auction for a patent crucial to a competitor's product, even if you paid somewhat more than the market value, you deprive a competitor of a technology or force them to license or cross
Re: (Score:2)
Could be. I think you also help make the case against software patents.
Re: (Score:2)
Re: (Score:2)
Why is bidding pi billion more foolish than bidding 3.15 billion or 3.10 billion? The aim is to bid slightly more than the second highest bid but less than the value to you of the item. They can only estimate.
Auctions are quite tricky things. The expected winning bid if everyone is rational and everything is ideal is the second highest valuation of all of the bidders in the room. That's true not just for a standard auction (bids going up, winner is the last one standing), but, on average, for weird kinds of
Re: (Score:3)
The expected winning bid if everyone is rational and everything is ideal is the second highest valuation of all of the bidders in the room.
Ah, but Google have demonstrated that they're transcendental instead of rational.
Re: (Score:2)
Patents have irrational value (Score:5, Funny)
All these bids are irrational numbers.
I think the message is clear. Patents have irrational value.
Re: (Score:2)
No, Google should have bid a complex number.
So they can split the value of patents between their real and imaginary values.
Also, no bid would be greater (or smaller) than that one.
Re: (Score:3)
No, Google should have bid a complex number.
I mean, they kind of did that with their bid of $3.14159 * 10^9 + 0i, right?
Just imagine what the accountants would have done if The Goog had tried to bid a non-trivial complex number.... probably just complained that they didn't have a column for "i" in their spreadsheet... :-)
Re: (Score:2)
No, Google should have bid a complex number.
I mean, they kind of did that with their bid of $3.14159 * 10^9 + 0i, right?
Just imagine what the accountants would have done if The Goog had tried to bid a non-trivial complex number.... probably just complained that they didn't have a column for "i" in their spreadsheet... :-)
These are Google accountants. After they finished determining that the quarter's numbers all lined up with Benford's law [wikipedia.org] they would have complained that the complex number violated type safety constraints and thrown an exception.
Re:Patents have irrational value (Score:4, Funny)
It gets better. If Google borrowed an amount of $ with a nonzero imaginary component, through the miracle of complex number exponentiation eventually the bank would owe *them* money.
Re: (Score:2)
You are of course assuming that the banks don't know about imaginary and complex number exploitation. How exactly do you think they got a profit out of the greatest financial crash since the Great Depression that they created *and* got bailed out to the tune of $700bn?
Re: (Score:3)
Apart from, perhaps, a larger complex number?
In which sense do you mean "larger complex number"?
Next bid? (Score:2)
Had they wanted to continue, would the next bid have been Feigenbaum's (4.669201609) billion?
Re: (Score:2)
or Levy's constant 3.275822918 billion
It must be fun... (Score:2)
Re: (Score:2)
Re: (Score:2)
to be so rich? it's investor money they were bidding with,
It's a mix of investor-provided cash from bond and stock offerings and liquid assets, I would imagine.
as such you'd like to read reasoning for that in quarterly reports, not bullshit about how they saved few tens of millions by their "fx hedging program" (no shit, they really think that's worthwhile info to spend couple of slides on and then just skip over the billion dollar stuff).
That's fairly substantial. Just because you don't want to hear about it, doesn't mean it's not important.
but did they even want to win the bids? did they have solid reasoning for the bids worth? did they even check what they were bidding on?
That's all been covered previously on Slashdot. This was a key strategic buy that was aimed at growing their patent warchest in order to survive patent challenges against technologies like Android (e.g. the Oracle/Java suit).
have you noticed how android is not actually letting fresh new players enter the phone market? niche production numbers aside, only the same old moto, samsung, lg, huawei, sony-e etc are in the game and they got their patents and licensing for the patents covered.
Uh... smartphones were the sole domain of 2 primary players and one "yeah, right
Big pies (Score:5, Funny)
That's a big piece of pi.
-Charlie
They have done these stuff many times (Score:3, Informative)
Re: (Score:2)
Don't be a douche.
Source: http://www.marketwatch.com/story/whats-your-google-bid [marketwatch.com]
WTF Google... (Score:2, Insightful)
We're bidding billions of dollars and we round Pi off at 5 decimal places?
They should have bid $3,141,592,653.59.
Re: (Score:2)
Weird! (Score:2, Offtopic)
When I read this article, there were exactly 42 comments posted!
Oh, damn....
Nerdy is cool but.. (Score:2)
You have to wonder if all the wires are plugged in when you make billion dollar bids based on math constants.
Re: (Score:2)
Principals didn't get it (Score:5, Insightful)
Re: (Score:2)
The only good people in society are in finance and top corporate management.
Why do you draw artificial distinctions between financiers and top corporate management?
Re: (Score:2)
Pi (Score:5, Funny)
Re: (Score:2)
If only they had bid Tau they would have won.
Re:Pi (Score:5, Funny)
I thought pie are square?
Great time to end software patents. (Score:2)
End software patents, leave this consortium with 4.5 billions dollars worth of paper.
In any case, I hope this deal fails to pass regulatory inspection. Apple monopolizing smartphones with patents is probably worth 10-20 billion to them. No one should be able to monopolize technology. The whole idea of patents made more sense hundreds of years ago when everything was trade secrets, but in the information age it's just a way to create artificial monopolies on certain technologies.
Re: (Score:2)
These weren't software patents, but related to telecommunications : 3g, 4g, networking, etc ... These would have been used defensively by Google to protect Android.
a deliberate ploy... (Score:3)
Re: (Score:2)
Uh-huh. Because the world's largest online advertising company needs to finesse a story onto Slashdot to gain public awareness.
Winning auction is not a win. (Score:2)
They probably could have bought these patents but obviously they are not worth that much to them. So their competitors paid for these patents more than Google was ready to pay for it, hence, as far as Google was concerned their competitors paid more than the patents are worth.
Bidding round numbers is a bad idea, you will learn it quickly on eBay (or maybe not since 99% items now are BuyItNow stuff that is cheaper on DealExtreme anyway).
Patent Trolls... (Score:2)
We all know what this means (Score:3)
Google has a sentient bidding algorithm and it's trying to communicate.
Re: (Score:2)
Well, actually that's an interesting question. I really wonder how much Apple and company were willing to go up to. It seems to me that Google and Intel together would have been worth their while pushing their bid a fair bit more; the Microsoft/Apple cartel was clearly pretty desperate and I think they might well have bid above Tau.
To absolutely guarantee a win, I think Google would have been safe with going for their namesake a Googol [wikipedia.org]. I doubt even Microsoft would be willing to outbid that, even if i
Re: (Score:2)
To absolutely guarantee a win, I think Google would have been safe with going for their namesake a Googol [wikipedia.org]. I doubt even Microsoft would be willing to outbid that, even if it was in Somali Cents. I have no idea why they didn't just go for it.
Hmmm, a Googol of Ostmarks, perhaps?
Re: (Score:2)
Well, it's not like they bid $3.14 for pi. They just need to divide Graham's number down by 10^x, where x is some unfathomably huge number, and solve for the first 12 digits. Should be easy for smart guys like Google, right?
...right?
Re:Of course they lost! (Score:5, Interesting)
I'm pretty sure it wasn't the math gods they were trying to provoke.
Instead it looks as if they were in a non-serious bidding game to make the others over pay for what are probably soon obsolete patents anyway.
Re: (Score:2)
The problem is that they should have been using Tau.
Re: (Score:2)
Re: (Score:3)
I think what Google was after was
a) some additional PR (and they got it as we see)
b) a nod from the nerd community which sometimes seemed a bit alienated lately (and they got it also as one can see from the above comment)
And they even got it for free (-lawyers' costs) as they didn't win the auction. So much for waisting shareholder's money as someone claimed above.
Re: (Score:3)
Re: (Score:2)
Re: (Score:2)
Maybe all those billions will end up having been paid for worthless software patents.
Correct, that would be the best outcome. I thought about this too, and the question that I have been asking myself is whether such a reform would be legal at all. I can imagine a reform that forbids NEW patents, but invalidating EXISTING ones retroactively? Wouldn't that be prohibited by some "no punishment without pre-existing law" meta-law?
Re: (Score:2)
Nortel makes hardware. Many (most) of those patents would be hardware patents, completely unaffected by software patent reform.
Re: (Score:2)
Unfortunately this may spell the end of "cheap" Android phones since the LTE patents would need to be licensed. Google should have been more serious, but this kinda smells like they may not be committed to the Android platform.
You couldn't be more wrong. The manufacturers get the required radio patent licenses either directly or with their radio chipsets (Broadcom, etc), not with the OS. This won't affect anything at all.
Re: (Score:2)
Umm... I can't imagine that number, how much is it in bailouts?
Re: (Score:2)
To be fair, it's not a unit that you'd need in everyday life, not even as a math prof.
Re: (Score:2)
why? it's a piece of useless trivia, notable only for being a "large" number (and it isn't that large, compared to some mathematically relevant numbers).
Re: (Score:2)
you didn't think of googol+1? ;-)
Re: (Score:2)
Re: (Score:3)
How about the opposite? It's a way to bid strange numbers without the possibility of being accused of collusive bidding. It's a standard strategy on auction sites to bid a strange ending. Newbies bid round numbers. People who want to win against newbies bid a bit more than a round number. People who want to win against people who want to win against Newbies bid a strange number.
If Google had bid more or less random strange numbers, then they could be accused of signalling. Since they bid obvious wel
Re: (Score:2)
Hahaha your whitespace-encoded message at the end took me a while to decode.
Very funny!
Re: (Score:3)
> Wait, what? You can buy constants now?
That was news to me, too. I thought you could only buy vowels!
You don't get it... They bid numbers for a reason (Score:2)
Re: (Score:2)
The majority of Nortel's patents would be hardware - being a telecomms manufacturer and all. Software patents are completely irrelevant.
Re: (Score:2)
The radio patents don't come with the OS, they come with the radio chipsets. This has nothing to do with 4G patent protection for the OS vendors.