HNHacker News
TopNewBestAskShowJobs

gmt2027

342 karma · joined April 13, 2011

submissionscomments
gmt2027··on Ask HN: What data structure does our brain use?
Actually representing the entire state of the universe sounds rather ambitious. However, the very action of 'imaging' will force the entire system into a consistent state by quantum collapse. The universe will be represented but changed, the original multitude of super-imposed states will be lost.

Heisenberg's uncertainty principle implies that some things are just unknowable. However, only the most basic constituents of physical reality demonstrate noticeable quantum behaviour. The macroscopic universe is fairly deterministic. You don't need to jump in front of a moving bus to learn that Newtonian mechanics applies 100% of the time in a very deterministic way.

gmt2027··on Ask HN: What data structure does our brain use?
I believe we are considering an abstract data structure here.
gmt2027··on Ask HN: What data structure does our brain use?
Short answer: A graph.

All data structures are simplified graphs.

The state of the physical universe is a massive graph in which interconnected objects are themselves massive assemblies of graphs of atoms and the atoms are graphs of subatomic particles. It's graphs all the way down. The properties of all systems - physical, chemical, economic, biological - emerge from the interactions between simple connected elements.

My opinion is that all knowledge is representable as a connected graph. The disconnect between our computers and our minds arises from the fact that brains are categorically not numerical machines but graph processing and pattern recognition engines. Neural networks are the underlying hardware and, with the typical elegance of nature, these are also graphs.

It should be possible to build a graph based language. The basic "Elements" [SICP] are easy to realise:

1. Primitive Expressions are graph nodes. They have identity and not much else.

2. Means of combination. Graphs can be added, subtracted etc.

3. Abstraction. A graph can be abstracted into a single node. We have no problem looking at a complex assembly of components as a single entity.

Since Google and Facebook are two massive platforms whose value arise from direct interaction with planet-scale graphs with billions of nodes, would these platforms be easier to build if our computers were more graph oriented? I would like to believe so.

gmt2027··on Why I do not want to work at Google
I agree, the discussion here has immediately taken a technical dimension for the same reasons that engineers are really poor at evaluating the real world value and far-reaching social implications of the technology they create.

This is not an article about the technical difficulties of running today's Internet or email. It is about the transition from revolution to institution. Centralisation limits the disruptive potential of the Internet and in the long term will transform it from an enabling technology into another obstacle that needs to be overcome.

This is just another step in a long series of advancements that have transformed human existence: organised Agriculture, Centralised Government, The Written Word, The Legal System, Mathematics, Science, Democracy, Free Markets, Finance, The Industrial Revolution, Publishing, Education, Modern Medicine, Personal Automobiles, and the Mass Media. In the idealistic minds of the pioneers, the potential was evident and intentions often subversive. Today these have invariably matured into terribly inefficient monstrous empires that are fiercely resistant to innovation or new ideas. With the resources to move the world, why should they move for anyone?

Personal Computing and the Internet are new and exciting, the gatekeepers remain engineers because they are the only ones that seem to be able to get it to do anything. Despite the convenience argument, building a centralised infrastructure will lead to the inevitable takeover by the same maniacs that run everything else. Of course, it may take a few generations before Google's superbly engineered infrastructure is operated by a Manufacturing/Pharmaceutical/Petroleum/Media/Retail/Finance/Defense conglomerate whose alumni occupy every significant position in Government. Why should we bother?

"After this, let's build the biggest, most bad-ass super robot weapon ever because it's cool and fun, we'll use a master password because it's more convenient and surely that bully that eats all our lunches might be watching intently, but he's too stupid to understand advanced control theory."

That being said, I would like to work at Google, the smart kids building the biggest, most bad-ass super robot weapon ever.

gmt2027··on How to get $12 billion of gold to Venezuela
"Catch them young"

Canada was a highly developed economy before the era of big oil. It has the third largest proven oil reserves after Venezuela and Saudi Arabia. Despite high production, primary industries constitute less than 6.2% of GDP. Contrast this with embryonic economies that were just emerging from colonial control, where a few million dollars was disproportionately above average income and you could buy yourself a government by creating a ruling oligarchy.

"We only negotiate with terrorists when they're selling oil."

The capitalist deception is that money bought weapons for the ruling class to do the dirty work of taking these resources by force when necessary. Doing big business with a thug empowers him; when he's several orders of magnitude wealthier than everyone in his community, you get a polarized society of highly successful armed gangs and oppressed people. When he's also a religious extremist you get the Taliban. Can a free market exist without a free society?

gmt2027··on How to get $12 billion of gold to Venezuela
Good article. It is rather depressing to accept the reality that these countries will only continue to deteriorate for the foreseeable future.
gmt2027··on How to get $12 billion of gold to Venezuela
I find this very easy to believe and it is the story of the poor major oil producing countries. In Nigeria, 25% of the national budget is dedicated to the personal expenses of the lawmakers by law. The rest makes its way into a very corrupt system that the multinational oil corporations are complicit in building after infiltrating the government and civil service at all levels [wikileaks]. They easily avoid accountability for persistent decade-long oil spills that exceed the Gulf of Mexico incident each year. In the long term, this 'oil wealth' is toxic to the society, infrastructure and environment of these countries.

The demand for oil is in every sense an economic addiction and the governments of the industrialised world will suspend all principle to keep the pipelines flowing. They empower ruthless sociopaths with billions of dollars and feign outrage when these despots hold their people hostage and slaughter thousands to maintain their hold on power. Why are the world's most dangerous men invariably from major oil producing countries?

I think the capitalist ideal of money as the only measure of success is a very irresponsible way to run a planet. In the past at least, the system seemed to work in favour of a few countries that were first at the controls. Now that there is just so much momentum that this ship has started to fly itself, I find it amusing to watch the US and Europe scrambling to come to grips with just a small part of the helplessness of the other 5 billion.

gmt2027··on Imperative vs Functional Programming
The basic problem in porting to OpenCL/CUDA is identifying the mapping and reduction operations buried in imperative code. That these data-parallel operations are fundamentally functional is clearly reflected in the primitives that the Thrust and CUDPP libraries provide.

The real challenge emerges during program restructuring and performance tuning. This requires intimate architectural knowledge and frequent trade-offs between conflicting optimisations.

gmt2027··on America Lacks Meaningful Innovation
Email already connects everyone on the Internet but billions of people are not online by an accident of geography and birth. Mobile phones connect a reasonable percentage of the rest. I may be mistaken but it is doubtful that many users' quality of life would be dramatically affected by the sudden demise of Facebook. It may be as bad as losing your favourite TV channel, but life goes on. Facebook is a convenience that does not solve any of our real problems or alleviate human drudgery in any way, It only increases the enjoyment of existing wealth at idle time.

Incidentally, an explicit statement of what mankind's actual goals are and what resources are being committed to solving them would go a long way to show that we really all want the same things. Anyone that believes our interests are best served by large corporations and governments at war is delusional. These entities rely on the premise that success can only come at the expense of the 'other guy'. In the grand scheme of things they are local optima that promote the very scarcity that they have evolved to manage. Open Source Software can teach something here, there are multiple distributions and programs, each freely borrowing from the successes and avoiding the failures of the other. The healthy but open competition between them means that we get the best operating systems that no money can buy. Even OSS is not immune to meaningless rivalries and sabotage.

The information age gives us new tools to approach the physical world which is where ALL the real challenges still are, even in computing (think the end of Moore's law, and the potential of quantum computing and communication). You mention mechanical engineering, we should have amazing open-source mechanical modelling software that realistically simulates the physical properties of large systems, allowing you to design, build and test machines from a catalog of reuseable components and subsystems in a virtual lab (Tesla did this in his head), and send it out into the cloud for fabrication and delivery.

We all seem to agree that life is hard, why can't we approach it as an engineering problem? I would like to see someone attempt to engineer a society in a systematic scientific way. Evolutionary algorithms, game theory and a few Petaflops of modelling power for the common objective function. No secrecy, sentiment or ideology just let the best algorithms decide.

gmt2027··on America Lacks Meaningful Innovation
We are a generation of fluff and polish.

Today's most celebrated young 'engineer' is Mark Zuckerberg, creator of a really cool way to rank hot chicks, measure faux popularity and extend the social dyamics of high school into the real world. We make dramatically scored movies about his trials and ultimate triumph and rank Facebook as the greatest company to emerge in the last half-decade with a $50bn valuation.

Where is our Alexander Graham Bell, Henry Ford, Thomas Edison, Nikola Tesla or even Howard Hughes? Is it Steve Jobs and his charisma? What we consider innovation has taken the form of the iPhone and the iPad, fancier toys in polished packages with glaringly less functionality than is a technical possibility today, conveniently dumbed-down so it is easier to keep us not thinking too hard and, God-forbid, doing anything really imaginative.

Sure there is interesting work being done out there in green energy, space, biotech and nanotechnology research. These were all conceivable decades ago. We are unable to create good science fiction anymore because our imagination is just as bad as it was in the 1800s. My God, they built driverless cars that work, why isn't that game-changing Google spin-off the hot new company of the decade? Does anyone even know the name of the former Stanford Professor whose work could lower accident rates, eliminate traffic jams and parking problems, make automobile ownership obsolete and drastically cut down emissions and manufacturing waste? We could have a real transportation cloud that actually does something useful other than being a repository for our videos and photos that allows 'sharing with family and friends'.

We are complacent enough to only care about things that distract us from actually being productive. Just about any system out there leaves huge room for improvement. Everything is broken or needlessly inefficient: the government, the legal, financial, energy, educational, healthcare, transportation, and disaster management systems. Even the Internet is broken. We should be building efficient sustainable systems that scale, not software. Real innovation requires an iterate-or-die mindset.

I am an African. Don't even get me started on the developing world.

← PreviousPage 4 of 4