Become a fan of Slashdot on Facebook


Forgot your password?

Comment OZ gov't is a bunch of whiners (Score 1) 179

The Aussie government failed to recommend a standard that supplants PDF in such a way that it handles all the cases one would expect to handle. So what's the point of this exercise that the OZ gov't did other than basically say without words... 'we should publish everything in XML documents since at least those can be parsed to some degree?

You know, there should be an industry-standard sheet of paper (Letter/AF) that meets the JAWS difficulty test, much in the same way there are test HTML pages that test web browser compliance with HTML 1.1/5.0.

Needless to say, blind people already have solutions for reading printed text that is not braille. Print the PDF and then scan it back into OCR-to-speech software. I'm sure someone by now has invented the OCR-capable print driver that eliminates the need to print to paper to reach the step of reading scanned paper.

Create a PDF document that has radially-printed text, "The green fox slept and fellated the brown dog." printed in a straight line, then printed in a spiral, and then printed upside down.

Then for Hebrew and Arabic (RTL languages), the same type of sentence... printed in RTL in various configurations.

Then the newsprint column layout, etc. etc. etc.

Point JAWS at the PDF, or use the PDF reader's built in speech interpretation, and let PDF vendors attain for certified compliance from the accessibility software industry.

Problem solved.

Comment Re:It's the Larry Ellison Parade (Score 2, Interesting) 66

I don't dispute that at all. SAP owes money. Question is how much. Did SAP really capitalize and make any gains off the back of Oracle that warrants a gigantic payout by SAP? That's what the trial is all about.

SAP did make some very smart buys by the way. BusinessObjects was probably their best purchase ever and probably was a huge thorn in Oracle's butt... and Leo was lucky to be present on the closing of that deal. Oracle has yet to come up with a decent fully-integrated BI suite to match what BO users are doing with that software. And now that BO is permanently married to SAP, Oracle has to steer their customers away from what is actually quite a nice product.

Comment It's the Larry Ellison Parade (Score 3, Interesting) 66

Too bad it's not a jury of dimwits that gets to determine the award. Then they'll have to ask themselves in a conference room if Larry's crocodile tears are worth excising a large sum of money from one of Germany's largest companies and giving it to a douchebag who already has loads himself.

Maybe if there's some scarlet twist to this suit that hasn't been made public yet----like SAP had intended to modify Oracle's code to make the database columns with German identifiers like SAP's?

I cannot tell you with what joy I have going into transaction SE38, then digging through ABAP code, then trying to figure out what column identifiers (like LOGIKZW, STABPRRT) means...

Comment Re:HFT is Not a Sin (Score 4, Insightful) 178

Another lie the chairs of NASDAQ and NYSE love to tell the world is that HFT speeds up price discovery.

How is this even POSSIBLY true??? They are algos. Those algos don't have any clue what the future performance of a company is. The algos are not going to tell you how successful AAPL's iPhone 5 will be, or when the next class action lawsuit is coming.

And algos break ALL THE TIME. It has been happening more often these days because stocks are breaking 120DMA more often, and most of these algos are doing nothing but backtracing trends on top of their arbitrage schemes. When a big investor comes in the room, they jump on him like nervous poodles.

That's why the May 6 event was such an eye opener. Waddle and Reed didn't cause the flash crash. They executed a normal transaction that wasn't even a Big Fish transaction, and all the algos went haywire.

So much for the quants and their MIT-smartness.

Comment Re:HFT is Not a Sin (Score 4, Insightful) 178

Yes it IS a sin.

It's a SIN because the heads of NYSE and NASDAQ continue to spread this lie that HFT shops contribute liquidity to the system. THEY DO ANYTHING BUT!

Have you seen the offices these HFT shops rent out in New Jersey and CT? They're cheap Class B space, warehouse loft and other low-rent space. They don't have the capital it takes to be a market maker. They just have capital---and they aren't going to sit in the market when it is hurting and make trades no sane person would make to keep liquidity flowing.

That is the job of a REAL market-maker. A market-maker will step in and be the counterparty to keep the issues they are responsible on the exchange moving.

What the fuck do you think happened on the May 6 flash crash? Almost all the HFT shops ran to their server rooms and SIGSEGV their software and pulled out to avoid taking more pain. The bids all dried up on the NYSE which is why the first crazy market order for $0.01 a share came to the exchanges, NASDAQ cleared it so for a while, several stocks were at zero print... like Accenture.

When you have the most top companies on your exchange printing zero in the flash of an eye... YOUR MARKET IS BROKEN. How is this even debatable?

Why would I want to put the kids college fund money in this fucking disaster?

HFT sucks.

Comment Re:Time for a rant... (Score 4, Informative) 178

Hear hear.

I pulled out 100K out of the markets because I can't just put up with HFT anymore. So Buy and Hold was a bad idea. Now investing is a bad idea.

You can't put in a stop-loss order anymore on anything you own because every day you have to worry if a mini-flash crash hit one of your issues and triggered it, then the SX won't unwind YOUR trade but they are glad to unwind the fuck-up trades the HFT guys caused.

Comment Re:A time out is the right solution. (Score 1) 218

Actually there WERE enough orders to satisfy the selling, the problem was that the order volume was mostly HFTs that were trading with each other in S&P e-minis. The algorithm was judging the liquidity of the market based on VOLUME, not based off absorption of the securities.

Absorption is when some buyers enter the market, buy stuff, then walk away and sell it at a much later time while everyone else in the market has moved on. They may come back a few hours later to re-sell now that the price has gone higher, but they're not recycling the same securities and getting them handed back to them on the very next trade.

Look at it this way from a programming point of view. You have 2 threads that are sitting on the side, and juggling the same balls between each other. The balls are worth $1,000 each and there are 100 of them being juggled.

So as the stupid hedge fund manager, you insert an order to sell off 5,000 balls at $999 a piece. The order fills for $950 and the two threads pick up the extra 5,000 balls. And now they are trading 5,100 balls between each other, and the price now stabilizes between $950 and $960 per ball. The new balls being passed around have less value because there's more of them in the market so there is dilution, but there is trading volume---which PORTENDS a normal investor to think there is liquidity, but there isn't---because there is no absorption and there's no sideline money... it's an empty market with a few players doing a massive amount of volume.

So take the same situation but now the stupid hedge fund manager starts POURING balls into the 2 threads. Now they are overloaded with volume and trading activity is higher, but there's no sideline money and the number of threads hasn't changed, so now the market has gone beyond saturated. The price per ball plummets.

The balls are linked to other objects which have value, and these start going down (we're talking futures contracts now). Because the OUTSIDE market is already heavily down, the futures that were being depressed triggered program trading---sell sell sell... cover your ass, don't get burned, don't ride this ship down. So now you have a full-on stampede.

NOW you have people that were holding the e-mini contracts that were asleep during this episode suddenly awake with alarm, and they start to sell because now their contract hedge is going down the toilet. You now have a perfect shitstorm of selling.

When Arca put on SLOWDOWN and the e-mini pit recovered after the order finished executing, then you had a stampede of buying from all the traders who saw immediate discounts on everything---everything is now on sale at Dollar Store discount bargain prices. So in a split second, it's now BUY BUY BUY!!!

And more than 50% of that snap recovery were all the folks and program traders trying to reverse their downside trades. They sold low---so they tried to grab all their positions back before they had to print a loss for the day.

In other words, a shitstorm of VERY smart people dumped into unloading a lot of holdings just because one simple program went "BOO!". They saw their collective mistake, and in a few minutes, they tried to reverse the damage so they wouldn't have to report losses.

While the episode went down though, a LOT of people in the tri-state area were having heart attacks.

Comment Re:A time out is the right solution. (Score 2, Interesting) 218

You use a market order if you have no time to sit and wait for your limit to trigger and you're very motivated to buy or sell to get rid of a stock or to pick a stock up off the floor.

Market orders rule when it's 3:59:50 and you WANT to get that trade done so you don't get exposed to overnight trading and the morning futures market.

Comment Re:Valuation is an art (Score 2, Insightful) 218

Sellers can also throw out non-nonsensical prices just like bidders can offer nonsensical asks.

That's the trouble with some of the algorithms that HFT shops are allowed to run in their boiler room datacenters. Some of these guys can only afford 80ms-400ms of time to create a decision, and that doesn't leave them much room to insert code that takes a 25,000-foot-view of their total order volume and determine if their order flow is making sense or not, dealing with 50,000 one-off situations (such as somebody dumping huge amounts of paper on the futures markets by accident)..

and so the quants who came up with the automation trading algorithms and the programmers who put their drawing-board ideas into software code can only put circuit breakers in their own code that are self-serving, but have no interest to put in any collars that stop the trades if it's clear that the trading is damaging the market.

It's really too late now. Retail investors have been bailing out and moving over to safe-havens and leaving the exchanges to hedgies and HFT shops to mutually-masturbate with each other trying to steal each others' money.

Comment Re:A time out is the right solution. (Score 1) 218

You can see the often-said argument that the Flash Crash caused Zero-Prints was due to insanely-low bid prices is not exactly accurate. Sure there was some crazy offers, but look at the trading volume.... it was all execution nonsense coming from the HFT trading systems that was flooding the markets with non-sensical orders.

I kind of wish the exchanges DIDNT reverse the trades. It would have wiped out more than 3/4ths of all the HFT shops and sent them packing. Now I won't even dare put any IRA/401(k) money in equities much less any in cash brokerage, unless it's a company with astronomical cash balances and pays a fat dividend--which is hardly nobody.

I won't invest in mutual funds anymore because there's no way a tired old fund manager can keep up with constantly-changing HFT strategy.

Comment Re:A time out is the right solution. (Score 1) 218

This isn't the true picture of what happened.

The pit where this drama went down was JUST the S&P 500 e-mini futures pit. Here's the tick by tick and audio of the pit boss having a heart attack over it. This was the pit where the fat-fingered trade happened. Actually, it wasn't exactly a fat-finger but a mistimed trade... it was supposed to take 5 hours to sell the futures off in small little bursts, but instead they "flooded the market with paper".

This spilled over into the REAL stock exchanges when the futures indexes all trended down, this cause the HFT guys (High Frequency Traders) who trade and/or watch the e-mini market to queue orders to start selling on the actual equity side, which already had shallow liquidity. This caused stocks to break through the 100, 150 and later the 200-day moving averages, which then triggered off the program selling.

NYSE at this point had already turned on their SLOWDOWN code which caused all the HFT computer systems to think that there was something wrong at the NYSE, so the order flow that wasn't going to NASDAQ and the tertiary exchanges to get the flood of sell orders.

Everybody who had collars on their stocks to limit-sell at 10% and greater lows then triggered off and a stampede to sell ensued and dropped some S&P500 stocks (like Accenture) to print $0.

The HFT guys were running to their servers and trying to shut everything down while the madness ensued.

So much for all these fancy high speed servers in New Jersey and Connecticut. All it has done is convince the retail investors to toss their money in Treasuries and forget about it.

Slashdot Top Deals

Outside of a dog, a book is man's best friend. Inside of a dog, it is too dark to read.