zygomega

I often think about functional programming as a directed graph consuming information.

The author addresses the potential benefits of immutability and laziness in a note on cover trees[0]. Last time I checked, this is still the world's fastest NN algorithm.

Haskell could shine in numerical computing. I think the key is polymorphism: a vector is a matrix is a tensor is a number, and they can all be multiplied in the same way, once you define what multiply is.

Poor support is more about the lack of haskell coders doing ML. It's hard to compete without the numbers that other platforms enjoy.

[0] https://izbicki.me/blog/fast-nearest-neighbor-queries-in-has...

Akerlof's: If you don't know whether you are being sold a lemon, you probably think you are, and will only pay lemon prices. Hence brands and hence public market information openness.

Minsky: stability begets hubris begets instability in an endless cycle of humanity's economic systems. Governments need big balance sheets to sort that shit out.

Samuelson: free trade with low-wage nations could hurt workers in a high-wage country. Duh.

Keynes: beyond a point, people will not borrow any more no matter the interest rate, thank you very much. When this happens, governments should borrow (at very low rates) and spend to avoid economic malaise and collapse. They should not be austere.

Nash Equilibrium: Watch Beautiful Mind

Mundell-Fleming: A country can only ever do two of these things: fix the exchange rate, decide on interest rates, allow the free flow of capital. It's amusing when they try and do all three, or pretend to be doing something else.

Fishing for them as a kid, they were notorious for playing dead, then jumping out of the bucket and waddling back down the bank. Smart fish.

Does anyone know why communication is being shut down? Can they leave it on so we can get telemetry?

I was at a meeting just yesterday, where I said, the state borrows when rates are low. Just like SimCity.

He's sending them out to die perhaps, in his name. Is he out there? Not a nice move.

The army is secular. Every 20 years they remove some dictator or zealot that the system seems to throw up regularly. I'm not sure if there's been a split in the armed forces before, but the police always cop a hammering.

It might start off with version control, but it would then end with us gulping plump helmet slurpees watching the latest cgi DF-Marvel tie in at the local megaplex. Leave the brothers be, and let future historians earn their pay.

I wish 'programming' included teaching our kids how to be proficient users of technology. How to effectively google, how to bring up a terminal, ssh to another machine, how to secure your privacy are the type of skills I hope are included in this definition. The architectural and process engineering aspects of programming are better left to learn in adulthood.

In practice, it's the opposite. Switching may be a sign that you are a bad driver but it's also a sign that you are price sensitive which is usually the more important signal. Never switch and you are flagged as price insensitive and therefore a fine candidate for being over-charged.

It's a grey area legally, and is usually prevented by ethics committees and industry bodies. Practically, if you collect the information you run the risk of a large anti-discrimination suit down the track. The usual proxy for race/religion is first name, last name and zip-code - name clustering gives very strong risk differentiation (I crunch numbers for insurance companies).

Rents can be seen as a future cash flow stream, at least partially protected from erosion by competition, that is a reward for past efforts. In a world where interest rates decline after the establishment of the rent, super-normal profits flow to the rentier. Keynes missed this in his 'euthanasia of the rentier' tirade.

Modern FP thinks of object oriented processes as comonadic. See http://www.haskellforall.com/2013/02/you-could-have-invented... for a good exposition.

> there is a reason we have labor productivity statistics that try to quantify this

Productivity measures have been pretty screwed since services became the dominant labour form. If my dog groomer decides to charge me an extra 20% then her productivity has gone up by 20% - she is now producing an extra 20% of services per hour.

> But with most jobs are already meaningless from the viewpoint of economic productivity

I have no proof of this at all - just an anecdotal gut feel. For every content producer where I work, we have 2 agile coaches, a LEAN coach, 2 marketing types (who are probably the really productive workers - promoting and protecting an established brand monopoly), 5 ops and dev supporters (in an unreachable call centre somewhere in the world), and then an endless pyramid of middle and upper management overseeing all this process. Within this context, no one every got a pay upgrade by volunteering to reduce the size of their empire.

And, also within this context, there is a lot of unhappiness that I put down to lack of meaningful labour output making the world a better place. Meanwhile, my leisure time productivity has exploded - I consume whatever media I want in whatever form I want whenever I want. I'm at the lower socio-economic end of my tiny middle-class ecosystem but my diet would be the envy of the kings of yesteryear.

My point being is that we already have too much stuff in the world, and the existing system of trading labour for putting food on the table just encourages more stuff. I would agree that we are far from pareto efficient and are, in fact, trapped in a scheme where scarce resources like primary goods and human time are wasted. But I have no idea what the solution might look like.

A major fault line in the current economic system is the concept of 'right to work'. The current structure presumes that the right to receive a proportion of economic goods of a society should be strongly associated with employment. But with most jobs are already meaningless from the viewpoint of economic productivity, the 'right to work' ethic concentrates power in the hands of the managerial classes in charge of job placement, who are very good at inventing new ways to pretend that this is all economically beneficial.

The truly horrendous problem with all of this is that our standard of living is statistically quite high so the system is sustainable for a long while, until it eventually collapses under it's own weight of pointlessness.

I disagree. The evidence in quotation, plot detail and story-line convinced me it was either Pynchon or a brilliant mimic. None else does it like this.

> To begin, I’m going to illustrate a mathematical fact.

'growth' rate is likely to be a geometric constant, not an arithmetic one. A 0% growth rate followed by a 6% growth rate is not 3% geometric growth on average. 100% growth followed by -100% growth isn't 0% growth on average.

Many a quant manager has gotten rich off of spruiking the reverse of this story.

The market price of capital is the discounted value of future production (which will be equal to consumption). If the discount rate declines, then the price of capital goes up and at least some of this effect finds its way into measures of capital growth (and capital return).

Windfalls accrue to the current generation of risk capital holders and, to some extent, the current generation of consumers. Losers are everyone else - current savers and future generations.

And ditto the reaction of the community. No aggressive denunciations or vindictive chat - just calm and helpful trouble shooting. +1 for the culture they have created.

Thanks Beagle, and I appreciate the sentiment and goodwill behind your comments. I remain optimistic, however irrational that seems. Big Finance is oriented to Big Data crunch-and-run processes as you describe. A highly non-linear, inclusive, non-backtested approach can't be even imagined within that paradigm. With respect to open sourcing,, well I don't think I personally have a choice. I can't do it all by myself, don't want to work on a 1000-core backtest platform. But I'm finding lots of talented hackers that share my views and want to give this a go.

I hope you're right! Cory Doctorow suggests the main problem for writers is obscurity not intellectual property theft and I think the same can be said about about traders. By the time we get successful enough for people to take notice so that the weight of money destroys the initial edge,, well I think that's a pretty happy place to find yourself. The trick is to find the next edge by then. What we're really banking on is that an open-source approach can obtain results way beyond what you can do initially in walled-off closed-minded intellectual property cubby holes. And then, that the economics of being involved in the open-source project contnue to be biased towards collaboration as opposed to cutting and running with a snapshot. The value isn't in the code - it's in the process that produces and continues to innovate the code.

I'm thinking of a much shorter time horizon. The problem with long-term investing right here and now is that potential possible futures include many where the long-run risk-asset investor ends up with a big fat zero. Someone, somewhere will have to cough up and accept massive losses to rebalance the economy post QE.

The graphviz dot language: http://www.graphviz.org/Documentation.php

It has good support in org-mode and a there's a nice haskell package that takes the dot code and turns it into an internal graph representation.

Yes, very loose terminology. I was thinking of a combination of two events:

1. VWAP based trading of large positions, which creates assymetric momentum effects in volume and price (and which is then somewhat forecastable)

2. Not then recognizing that you are forecastable (as a result of playing the VWAP game). The extra information that someone who looks at intraday price relationships has over someone who doesn't. If you wade into the middle of a market that is short-run forecastable (eg it's trending downwards to a new level and the market maker/HFT guys are battling their battles), and you don't check as to whether it's short-run forecastable, then you're probably the patzy at the poker table.

That's why I thought the DNQ iqfeed was a suitable choice - no consolidation at all. Would love to know what the "interesting" effects are (beyond the fixed book CME effect) - will be looking a bit harder at the stream to find them!

Extremely helpful - this really clarifies the differentiation in the Julia approach and I'm excited your team is taking this direction. There's a lot of people cheering Julia on, even if we're wimping ut remaining on the sidelines.

It would help if this explanation was a bit more prominent.

I get the sense that the Lodestone project is now defunct.

Not having to register and login is one immediate differentiator. Marketcetera is a big, big code base - I think our ambitions are more focused.

It does feel like I'd be swimming against the tide heading down the R path. Thanks for the offer :)

It's not specific, it's just my recent experience. Hadley released his bigvis package a few weeks ago, and it contains the beginnings of a weighted average routine I sure can use.

mmap is a package that let me create a fast db from scratch in a day. ggplot, lattice and all the charting routines are a joy to work with.

I'm interested in random forests today, and I bet there's a package that links to the standard libraries whatever they are. This is also probably true for python, but Julia you'd have to build the API.