They will seperate it. Exchange members = no tax. Non-exchange members = taxed. The gap will continue to widen.... So if you're all complaining, why don't you start calling up GS, MS, BoA, and others to hire you? Oh wait... you don't have a track record or any real industry connection/exposure. Funny shit.
Huh? For example, before the advent of high technology, women were at a severe disadvantage in going to war. When sitting behind an F15, the differences amongst men and women are almost zero. Same thing with race car driving (look at Danika). In trading itself, when I was the MERC floor around 1985, there may have been five women on the entire floor as traders. Now there is maybe one in ten when screens became more prevalent. Before DSL, getting realtime data for the retail trader was a nightmare. Now everyone has it. With FIOS and networks like it, it may take 10 years, but I expect everyone will have multicast data right from the exchange for a nominal fee. Before the invention of glasses, I would be legally blind in my left eye. With glasses no big deal. Before the invention of the printing press, knowledge was in the hands of the clergy. After, Universities sprang up all over Europe. The list is endless.
Overall, I doubt that most stay-at-home traders are fast enough to be affected by 0.3 seconds. They usually lose because they have no edge, are overleveraged, are overtrading, and have poor money/trade management skills.
Ok I finally found out where some of the matching engines are: NYSEâs 55 Water Street and Weehawken facilities are home to the BATS and ARCA trade matching engines. ARCA has a backup in Chicago. As I figure out more I will post it. It is important to distinquish between data distribution centers, like Equinix, and the physical location of the matching engines. For example, Equinix NY4 center is in Secaucus, New Jersey.
This is why day traders are a dying breed, especailly scalpers. Learn to automate, or else your gone.
I am not disagreeing with you, but I think you go too far. You must engage in trading that computers are no good at. I hint at it in this thread in this post in response to the post above it: http://www.elitetrader.com/vb/showthread.php?s=&postid=1013611&highlight=Pragmatic#post1013611 Why aren't computers great at theorem proving? You would think they would be. Here is a plausible argument, and it has a great deal to do with what you are hinting at: http://ase.tufts.edu/cogstud/papers/penrose.htm Somehow, great human traders that trade on the shortest time frames must (are able to) read momentum by "non-computational" means in the way argued by Penrose in that article. It is a mixture of heuristics, computation, and pure instinct. In effect, these traders are/were using "Flash" orders "by hand" (probing markets with small orders and then putting on big orders once they assesed short term supply/demand imbalance) to discover liquidity, and then leaning on it. The natural progression was to automate what traders do by hand, and now you can run this algorithm almost as well (better ?) than HF human traders on hundreds or thousands of symbols simultaneously.
DirectEdge, as well as the FX matching engines BrokerTech, LAVA and HotSpot is in Secaucus NJ, which note is close to the Equinix NY4 center given above (same building ?).