crypto5

I highly doubt Google/Baidu will get into business of recognizing manufacturing defects in fidget spinners or analyzing sensory data

They will build platforms for:

- IoT - single robotic KIT/SDK, which allows you to easily install sensors and integrate data into rest of the platform

- Cloud ML - image recognition/models optimization - will allow to train model and detect defects from previous step by two mouse clicks

and take a lot of added value from small companies.

Good news is that postgres is going to get multi-master replication soon in postgres 10

Where did you read this?..

It is a lawyer's fees industry's size, not a legal software industry's size..

Do lawyers usually send very detailed itemized bills?

Is this project free for commercial usage?

They move much less, and have much more fat and maybe something else in them.

But "datacenter systems" are only 170B out of 3.5T there..

It depends. You can store 4TB on single HDD, but reading and processing it can take many hours, so you may want big data stack to have your task paralleled.

In random cloud provider you may not get FS with compression on your machine..

Part of ecosystem. See how firebase today is integrated with analytics/ads.

I really want a statically typed Python.

Can golang fill this role?

And what are the specs for this node? CPU, RAM?

How would you compare your product with https://www.blazegraph.com/, which has similar feature set, but is much longer on market, and has wider adaptation(e.g. powers wikidata)?

What about support at the driver level for Java and .NET applications or heterogeneous database schemas?

I guess MariaDB JDBC driver already has two phase commit logic, because MySql has it for decade. It is enough for distributed transactions in "heterogeneous database schemas".

Will this cause double-buffering (one buffer for host FS and another for BTRFS) with doubled RAM consumption?

Is this the same like following?:

Future<Long> f = executor.execute(() -> myFunc());

Otherwise, Google owns any IP created while you're at the company.

Not necessary in California: Any provision in an employment agreement which provides that an employee shall assign, or offer to assign, any of his or her rights in an invention to his or her employer shall not apply to an invention that the employee developed entirely on his or her own time without using the employer s equipment, supplies, facilities, or trade secret information. http://law.justia.com/codes/california/2011/lab/division-3/2...

There are exceptions of course, major one: you have not compete with your employer.

CORBA was superior in terms that it had distributed two phase commit protocol, which gRPC doesn't have, and which is very critical in banking/finance industry.

You need to get your company to 10k USD monthly product revenue within three months.

You probably need to add how much you can spend on customers acquisition to your equation.

How many Musk-like or Brin-like individuals have come through TCS or Wipro or Infosys on a H-1B so far?

You don't know, these people are forced to be in underpaid H1B slavery for decades and are not be able to use business opportunities.

You can use coding as a kind of meditation to clear your mind from negative emotions.

over systems

What kind of systems?

If the supply of axioms and inference rules was infinite, with no finite alternate encoding, it would mean Godel numbering would fail because the process of assigning numbers to each axiom and rule would never complete.

I strongly disagree with that. There is no evidence of that.

And if L1 happens to be strong enough to encode Peano arithmetic, you can construct a statement a Godel statement G1 which refers to logic L1.

As I said before, Godel proved that your G1 is unprovable in specific framework of Principia Mathematics, which is first/second/higher theory/logic (consist on quantors, functions, predicate, variables, rules).

I don't see evidence that there can be no other framework even with Peano arithmetic inside which can't deliver consistent theory. You started speculating about framework with infinite set of axiom, and I disagreed with that, but there can be other type of infinite framework, say when you allow proofs of infinite length, then Godel numbering may be impossible, and it can be example of such framework.

Goedel's Incompleteness Theorem is parametric over formal systems, with first-order Peano Arithmetic being one of the weakest, most standardized systems in which it applies.

It is parametric over formal systems described in Principia Mathematics. That's it. It doesn't take into account other possible types of formal systems. At least I didn't notice this when reading actual proof.

That operator is called a Turing Oracle

I think it may be very different thing. I just gave you a quick example. That operator can be something very different. You can set measure on space of proofs, and derive concept of asymptotic proof, and say if proof is asymptotic, then it is proof. There can be many variations around possible formal systems.

and it's physically impossible

This is very strange argument. Turing machine contains infinite amount of memory, and likely is physically impossible.

See my comment about what do I mean: https://news.ycombinator.com/item?id=14039104

Unless you want your logic to be infinite, of course. In that case, the method would break down.

Godel numbering is infinite, because it is just natural numbers. Nothing prevents you to build the same proof for infinite number of axioms, rules, functions, variables..

Let me give you example. At first you have predicate calculus, and Godel completeness Theorem.

Now you add new tool: existence predicate, and you got first order logic, which allows you to prove Godel's incompleteness Theorem.

What is the guarantee exactly that more advanced systems can't exist? Say system with new 'quantum hack' operator. You can't prove formula from Godel's proof? 'Quantum hack' under some conditions covers missing gap in proof path by building continuum truth table and give you tool to check if formula from Godel's proof actually provable, or it is false.

Proof was made within specific framework, with very specific quantors and type of inference, I still don't understand how it prevents existence of more powerful frameworks with different quantors and inference where incompleteness theorems wouldn't work.

I always was wondering: Godel proved his theorem regarding formal systems described in Principia Mathematics. In my understanding this is some higher level logic with recursive functions.

But how this can be extended to the whole universe of all possible formal systems? Who guarantee that there will be no some new system with quantum-oracle-operator, which will not be affected by incompleteness theorem, and can self-proof self-consistency?

Even well known m-recursive functions (which are essentially Turing machines) are wider class than primitive recursive functions used in the proof..

Planet Money's story about why is that: http://www.npr.org/sections/money/2014/07/23/334076398/episo...

Did you just say we need a $1.5 trillion dollar fighter to fight goat herders? ISIS doesn't have an air force.

ISIS potentially can have/get manpads, and MANPADS can have big advancements in foreseeable future. Also US can face contested environment in operations similar to Kosovo war, operations in Libya, Iraq, Panama, where local army had more advanced SAMs in possession with needs to suppress them.

We literally have zero need to protect power across the globe. Our presence in the middle east is extremely costly and counterproductive to our security.

This is question of policy, and is a mess for me. I am not going to speculate about this.

And the F-35 is guaranteed to be terrible at close air support. No F-35 pilot is going to slowly loiter their supersonic within visual range of their targets and without doing that they can't see exactly what they are shooting.

The main weapon for ground support planned to be SDB2. Also optics/targeting system is presumably much more advanced on F-35 comparing to A-10.