Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror

Comment: Re:Super-low latency trading Considered Bad... (Score 1) 157

by skubeedooo (#37508084) Attached to: Low-Latency Network Shaves Milliseconds from UK-Asia Traffic

None of that profit the mega-corp makes is economically useful activity (ask any economist) - they're simply leeches in the system, able to legally game it to their advantage (and, to everyone else' disadvantage, since trading is a zero-sum game).

I broadly agree with you, but it's not true that there is no useful work being done. The arbitrage that this high speed link enables has the direct effect of bringing HK prices in line with London prices and Wall Street prices, with lower latency. It is a benefit to each regional market that you can get the best price in the world at the click of a button and not just the best price on a particular exchange. It is of special benefit to smaller exchanges as HFT arbitrageurs transfer liquidity from the large exchange to the small exchange.

Comment: Re:Monetisation will work and advertising will die (Score 1) 444

by skubeedooo (#37165986) Attached to: Ask Slashdot: What Will IT Look Like In 10 Years?
Currently it costs around $0.01 to make a btc payment, and about 30 mins to process. BTC devs estimate (don't have the link handy) that in its current form it could maybe scale to handle the number of transactions now processed by VISA/Mastercard, but of course it would need to scale way beyond that to handle single cent page view payments. It really is not a good technology for micropayments, much better for medium to large transactions IMO.

Comment: How do i keep up with technical material (Score 1) 337

by skubeedooo (#36905874) Attached to: How Do You Keep Up With Science Developments?
I once did a PhD in theoretical physics and I'd like to keep up with important new developments. I always found that it was better to read the actual papers rather than some leaky abstraction in a blog, but now I have another job I don't have time to look through *all* the different articles and trawl through the citations.
So, my question is this...is there some kind of pagerank or suggestion service for reading articles on arxiv?
I'm thinking something a bit like reddit, where it maintains a global score of each article as well as tracking what you read so it can suggest what to read next based on what you liked. Citations don't quite do this as they only point backwards.

What is now proved was once only imagin'd. -- William Blake

Working...