In truth, Google is not a technology company. Really. HP, Sun, Oracle, Microsoft, Dell etc are technology companies: people pay them for products which are the fruits of research and development.
Google is not a technology company. Google is an advertising company with a sideline in email hosting. That's where their money comes from.
If you look at the technologies you listed, with the exception of Java, almost none of them was made profitable by the company that invented them. I don't know why companies who can afford really serious, advanced "blue sky" R&D so frequently fail to commercialise it, but it's really common.
I have a feeling I will be the designated baddy for today's thread:D
I am actually a big believer in research spending, and I think that any company with above-normal profitability is mad not to do a lot of it. But there's a difference between "research" and "entering every and all market segments you can hoping that one of them will be profitable".
Basically Microsoft and Google are almost totally reliant on single lines of business (Office + Windows vs AdSense, respectively) for their profits.
Because they're not paying *any* of that money to shareholders, there's no incentive to economise. More to the point, they suck up innovators and lock them up in a structure where they're beholden to internal process and not able just to say "fuck it, this idea is awesome, let's sell it!"
Google are already turning into Microsoft on this front too. Small companies regularly out-innovate (I hate that word too) them. So Google just buys them out.
I think that refusing to pay *any* dividend is just control-freakery. And it's bad for the economy because it encourages speculators to buy on the basis of short-term share price fluctuations. It used to be that you looked at the fundamentals of a company, then bought and held onto the shares in order to get dividends. Now you buy and flip it because paying dividends is old fashioned.
Google have the same problem as Microsoft: they're too successful. They have a river of cash flowing through the front door and an allergy to paying dividends to shareholders.
Thus they're pursuing what I call the "spaghetti cannon strategy". They blast buckets of spaghetti up against the wall and hope that some of it will stick.
Eventually any such company becomes large enough that it cannot coordinate what the various bits and pieces are doing. The self-cannabalising overlap of Android and ChromeOS is a symptom of the spaghetti cannon working overtime.
Because god forbid you send any of the profits to the people who paid money to own part of a wildly profitable company.
As an Australian I have had to endure years and years of idiot Senators embarrassing me in front the world's nerds here on Slashdot. Finally we have a half-smart one.
I'm not an American lawyer so I can't be sure, but in general the common law works similarly in most countries on civil matters like contracts, trusts, equity and the like. In particular, in equity (and trusts are a major subfield of equity) the law is not applied rigidly. In equity Judges are compelled to act on goals of fairness, not on precise rules per se.
The rule against perpetuities arose out of the history of "Uses", which is an older form of trusts. Landlords would deed land to lawyers or friends "with such and such portion for the use of my eldest son Harold, such and such to the use of Edward" and so on.
There was no way to break these, and so over the course of centuries, the land holdings would become impossibly and uneconomically fragmented. Even if circumstances had radically changed, it could mean that the wishes of your great-great-great-grandfather was basically making you destitute.
Basically that's how the rule against perpetuities came about. Remember, that while a corporation has an indefinite lifetime, it is definitely *alive*. It reacts dynamically to changing circumstances. It continues to exchange title to goods, purchase or provide services etc. Companies "die" during windups or bankruptcies.
Imagine if companies that went bankrupt in 1650 had perpetual trusts on their assets. Whole parts of the world would be unusable because they were "only" to be used for growing cotton or whathaveyou. That's the kind of situation the old "uses" led to.
It's based on the person identified by the title, not the title itself. So if I used that form today, it would mean Charles, son of Elizabeth the Second, regardless of when he ascended to be King.
I know that as geeks we give lawyers a lot of putdowns for being dicks, but the "equity" branch of law is sensible. And nobody hates people trying to game the system more than a judge.
"To me transferring wealth is not the issue. This can most likely be done through some sort of managed corporation (for instance, create and well-capitalize a corporation who's sole task to is to support your frozen ass. That is the stated purpose of the corporation within its charter. Assign a bank as director of the corp. etc)."
That would be unlikely to work, IMO. Remember, trusts come from the *facts* of a situation, not merely from documents. A court could look at this scheme, reasonably conclude that it's a trust (someone holding common-law legal title to something with future-you as the beneficiary) and apply the rule against perpetuities. It's not as though people haven't tried for centuries to break that rule and failed.
" Being technically dead, any contract you had has long since ended and is unenforceable."
Which is another reason why the law of trusts would apply, not the law of contracts. But there would still be difficulty in assigning yourself as beneficiary post-death. You could make it a class trust (ie, "For the benefit of persons having this particular DNA..." or "For the benefit of persons knowing this 16 digit password"), but that would probably get shot down as too general, or perhaps, too obviously specific.
Trusts are hard to game because they spring from equity, and equity is not a rigid system of logic like a computer program. It is specifically flexible and based on judicial discretion to slap down people who try to game the common law.
Before coming to my senses, I used to be a law student. Trusts, including estates, was probably my favourite subject. The main vehicle for transmitting wealth between generations is trusts, because they are reliable and well-understood.
However, in Australia, and other common-law countries such as the UK, Canada and the USA, trusts have a limited life-time. The basic principle is that the dead cannot rule the living. It's called the "rule against perpetuities". If trusts could last forever, more and more of the world's resources would be tied up in trusts with narrow aims and the eventually all the world would be divided between trustees and beneficiaries. So goes the argument, anyhow (this is different from conditional gifts and foundations, by the way, before you start yammering about scholarships and charitable organisations).
The lifetime of a trust is specified at its creation. In the old days you could make it $DEATH_DATE_OF_SOMEONE + 21 years. So you'd have stuff like "For the life of the Prince of Wales and 21 years", the theory being that it's easy to know when the Prince of Wales carks it. More recently, most jurisdictions have introduced legislation allowing an optional ability to simply fix some time period, usually up to 80 years.
And that's the problem. If you go into cryo-storage for 81 years, then on awakening you may find that your trust was dissolved and the benefits distributed to your descendants. And until it's proved that you can really come back from death via cryogenic storage, I'd be amazed if the courts changed their stance. Because too many people would try to break the rule against perpetuities by being "frozen".
Of course, IANAL, this isn't legal advice, YMMV yadda yadda.
> Much more money is being spent every year on improving x86 chips than all the competitors combined.
By your logic, General Motors should be crushing Ferrari in the supercar market. After all, GM spends much more on their car development than Ferrari does.
It *was* MIPS. They licensed core designs and used the MIPS ISA. But they had a custom 6-to-a-die design with some specialist MPI and fabric circuitry of their own design.
>This was for 8 cores on the same motherboard by the way, if you are spawning MPI jobs over a network socket, expect much much worse.
They thought of that in the SiCortex design. They use a custom internal fabric to reduce the maximum hop distance to 3 between any pair of chips; and furthermore they included on-die MPI-handling logic to take that overhead out of the CPUs.
They didn't die because their customers abandoned them for something cheaper. They died because they had a cashflow crisis due to investors pulling out of a planned round of fundraising. They had millions of dollars of sales in the pipeline.
The lesson isn't "Don't compete with Intel", it's "When you run out of money, you're out of business". Or perhaps, "The financial crisis killed lots of otherwise sound businesses". Luck, as the OP pointed out, played a large part.
That's what they designed: it's basically a bog-ordinary Linux-with-MPI cluster in a box. They had a custom internal fabric that was far more efficient than ordinary switches and even had on-die MPI accelerators. They also shipped with compilers for C, C++ and Fortran.
It was meant to be a drop-in replacement for room-sized clusters for a fraction of the space and heat. Basically what killed them was cashflow.
Slashdot Mirror
In truth, Google is not a technology company. Really. HP, Sun, Oracle, Microsoft, Dell etc are technology companies: people pay them for products which are the fruits of research and development.
Google is not a technology company. Google is an advertising company with a sideline in email hosting. That's where their money comes from.
If you look at the technologies you listed, with the exception of Java, almost none of them was made profitable by the company that invented them. I don't know why companies who can afford really serious, advanced "blue sky" R&D so frequently fail to commercialise it, but it's really common.
I have a feeling I will be the designated baddy for today's thread :D
I am actually a big believer in research spending, and I think that any company with above-normal profitability is mad not to do a lot of it. But there's a difference between "research" and "entering every and all market segments you can hoping that one of them will be profitable".
Basically Microsoft and Google are almost totally reliant on single lines of business (Office + Windows vs AdSense, respectively) for their profits.
Because they're not paying *any* of that money to shareholders, there's no incentive to economise. More to the point, they suck up innovators and lock them up in a structure where they're beholden to internal process and not able just to say "fuck it, this idea is awesome, let's sell it!"
Google are already turning into Microsoft on this front too. Small companies regularly out-innovate (I hate that word too) them. So Google just buys them out.
I think that refusing to pay *any* dividend is just control-freakery. And it's bad for the economy because it encourages speculators to buy on the basis of short-term share price fluctuations. It used to be that you looked at the fundamentals of a company, then bought and held onto the shares in order to get dividends. Now you buy and flip it because paying dividends is old fashioned.
Google have the same problem as Microsoft: they're too successful. They have a river of cash flowing through the front door and an allergy to paying dividends to shareholders.
Thus they're pursuing what I call the "spaghetti cannon strategy". They blast buckets of spaghetti up against the wall and hope that some of it will stick.
Eventually any such company becomes large enough that it cannot coordinate what the various bits and pieces are doing. The self-cannabalising overlap of Android and ChromeOS is a symptom of the spaghetti cannon working overtime.
Because god forbid you send any of the profits to the people who paid money to own part of a wildly profitable company.
Did the ancient Greeks and Romans have Ruby VMs that are just now bursting back onto the cultural scene?
In most common-law countries, education is defined as a charitable purpose. So the *school* is a charity, not necessarily the sect that runs it.
As an Australian I have had to endure years and years of idiot Senators embarrassing me in front the world's nerds here on Slashdot. Finally we have a half-smart one.
Don't get too excited. Xenophon is in the same Senate as that oxygen thief Stephen Conroy.
I'm not an American lawyer so I can't be sure, but in general the common law works similarly in most countries on civil matters like contracts, trusts, equity and the like. In particular, in equity (and trusts are a major subfield of equity) the law is not applied rigidly. In equity Judges are compelled to act on goals of fairness, not on precise rules per se.
I'll have to defer to you on the mechanics of Delaware corporations. Well outside my limited studies.
The rule against perpetuities arose out of the history of "Uses", which is an older form of trusts. Landlords would deed land to lawyers or friends "with such and such portion for the use of my eldest son Harold, such and such to the use of Edward" and so on.
There was no way to break these, and so over the course of centuries, the land holdings would become impossibly and uneconomically fragmented. Even if circumstances had radically changed, it could mean that the wishes of your great-great-great-grandfather was basically making you destitute.
Basically that's how the rule against perpetuities came about. Remember, that while a corporation has an indefinite lifetime, it is definitely *alive*. It reacts dynamically to changing circumstances. It continues to exchange title to goods, purchase or provide services etc. Companies "die" during windups or bankruptcies.
Imagine if companies that went bankrupt in 1650 had perpetual trusts on their assets. Whole parts of the world would be unusable because they were "only" to be used for growing cotton or whathaveyou. That's the kind of situation the old "uses" led to.
It's based on the person identified by the title, not the title itself. So if I used that form today, it would mean Charles, son of Elizabeth the Second, regardless of when he ascended to be King.
I know that as geeks we give lawyers a lot of putdowns for being dicks, but the "equity" branch of law is sensible. And nobody hates people trying to game the system more than a judge.
They have the trifecta of crummy website behaviour: excessive pagination, click-through ads and lazily regurgitating other people's content.
"To me transferring wealth is not the issue. This can most likely be done through some sort of managed corporation (for instance, create and well-capitalize a corporation who's sole task to is to support your frozen ass. That is the stated purpose of the corporation within its charter. Assign a bank as director of the corp. etc)."
..." or "For the benefit of persons knowing this 16 digit password"), but that would probably get shot down as too general, or perhaps, too obviously specific.
That would be unlikely to work, IMO. Remember, trusts come from the *facts* of a situation, not merely from documents. A court could look at this scheme, reasonably conclude that it's a trust (someone holding common-law legal title to something with future-you as the beneficiary) and apply the rule against perpetuities. It's not as though people haven't tried for centuries to break that rule and failed.
" Being technically dead, any contract you had has long since ended and is unenforceable."
Which is another reason why the law of trusts would apply, not the law of contracts. But there would still be difficulty in assigning yourself as beneficiary post-death. You could make it a class trust (ie, "For the benefit of persons having this particular DNA
Trusts are hard to game because they spring from equity, and equity is not a rigid system of logic like a computer program. It is specifically flexible and based on judicial discretion to slap down people who try to game the common law.
Before coming to my senses, I used to be a law student. Trusts, including estates, was probably my favourite subject. The main vehicle for transmitting wealth between generations is trusts, because they are reliable and well-understood.
However, in Australia, and other common-law countries such as the UK, Canada and the USA, trusts have a limited life-time. The basic principle is that the dead cannot rule the living. It's called the "rule against perpetuities". If trusts could last forever, more and more of the world's resources would be tied up in trusts with narrow aims and the eventually all the world would be divided between trustees and beneficiaries. So goes the argument, anyhow (this is different from conditional gifts and foundations, by the way, before you start yammering about scholarships and charitable organisations).
The lifetime of a trust is specified at its creation. In the old days you could make it $DEATH_DATE_OF_SOMEONE + 21 years. So you'd have stuff like "For the life of the Prince of Wales and 21 years", the theory being that it's easy to know when the Prince of Wales carks it. More recently, most jurisdictions have introduced legislation allowing an optional ability to simply fix some time period, usually up to 80 years.
And that's the problem. If you go into cryo-storage for 81 years, then on awakening you may find that your trust was dissolved and the benefits distributed to your descendants. And until it's proved that you can really come back from death via cryogenic storage, I'd be amazed if the courts changed their stance. Because too many people would try to break the rule against perpetuities by being "frozen".
Of course, IANAL, this isn't legal advice, YMMV yadda yadda.
I'd be surprised if the whole field isn't absolutely blanketed with patents by IBM. Mainframes have this since the 80s or 90s, I think.
There was a business case. There were plenty of customers. But there was no cashflow. That's what killed them.
Yep, purely a business failure due to the crappy timing of the GFC, rather than market trends per se.
What really pisses me off is that Sun bought MySQL for a billion dollars when they probably could have gotten SiCortex for a fraction of that.
> Much more money is being spent every year on improving x86 chips than all the competitors combined.
By your logic, General Motors should be crushing Ferrari in the supercar market. After all, GM spends much more on their car development than Ferrari does.
It *was* MIPS. They licensed core designs and used the MIPS ISA. But they had a custom 6-to-a-die design with some specialist MPI and fabric circuitry of their own design.
Apologies, the above comment seems to have been assigned to the wrong OP.
>This was for 8 cores on the same motherboard by the way, if you are spawning MPI jobs over a network socket, expect much much worse.
They thought of that in the SiCortex design. They use a custom internal fabric to reduce the maximum hop distance to 3 between any pair of chips; and furthermore they included on-die MPI-handling logic to take that overhead out of the CPUs.
They didn't die because their customers abandoned them for something cheaper. They died because they had a cashflow crisis due to investors pulling out of a planned round of fundraising. They had millions of dollars of sales in the pipeline.
The lesson isn't "Don't compete with Intel", it's "When you run out of money, you're out of business". Or perhaps, "The financial crisis killed lots of otherwise sound businesses". Luck, as the OP pointed out, played a large part.
By your logic, General Motors should be crushing Ferrari. After all, GM spends much more on their car development than Ferrari does.
That's what they designed: it's basically a bog-ordinary Linux-with-MPI cluster in a box. They had a custom internal fabric that was far more efficient than ordinary switches and even had on-die MPI accelerators. They also shipped with compilers for C, C++ and Fortran.
It was meant to be a drop-in replacement for room-sized clusters for a fraction of the space and heat. Basically what killed them was cashflow.
That Sun pissed away one billion dollars on MySQL instead of buying out SiCortex. Smart move, Sun!