That was "Pascal greater than C" using the greater than sign, but apparently slashdot can't escape that properly....

If you don't understand, I won't String you along, but give you 256 good reasons ...

you keep using that word. i don't think it means what you think it means.

You win the Internets!

Sorry to respond to myself, but to add ...

32 MB seems like a nice sized "chunk" to be an image of an on-die cache.

The size of the file is so conveniently close to a power of 2 to suggest to me that this is an image of RAM. (L2 cache?)

Again, I'm suggesting that a snapshot of the core rules generated by a learning system would be more useful than the framework it would run on.

Do you have any case/statute cites for that?

(Not asking to be a dink, but out of curiosity.)

Let's assume GS's trading system uses learning algorithms to generate new rules. This is a very logical assumption unless you want developers tweaking the code year after year, day after day.

In that case, it's not the code that generates the rules, nor the data it uses to generate the rules, nor the engine that runs the rules and interfaces the markets that is valuable: its the current state of the optimized rules themselves that has immediate exploitability by criminals.

If you know GS's system is going to react to market behavior that does X by doing Y, then you can exploit that to make money off it to GS's detriment.

And an interesting twist. The accused referred to the code as "open source." That might indicate that he thinks it isn't copyrighted. If the rules [code] were generated by software itself, such as a learning system, there is a colorable argument that there isn't the necessary quantum of authorship for copyright under U.S. law. Naturally this wouldn't encompass other claims, like those based in trade secrets, employment Ks, etc.

To recap, the state of rules generated by the learning system would be far more valuable [and dangerous in enemy hands] than the underlying engine and database of historical data. In fact, those rules would be the crown jewels.

I'd add the above is 100% pure speculation based on inferences and some expertise in the area. (No inside knowledge of GS.)

Found a post on ACM by someone with same name as the accused. Looked like a person with research background in Neural Networks. No idea if it is the same person, but it would be intriguing to me if Goldman Sachs was using neural networks for trading.

One interesting facet: if two or more counterparties in a market had neural networks that were trained to coordinate and cooperate in ways that would violate trading rules (e.g. like bridge players sharing info through actions), would the company be liable if the neural networks had developed these exchanges by themselves? In other words, would it be an instrumentality for violating the law if it learned, on its own, to violate the law, and the programmers / administrators "had no idea" it was doing that?

I live in the midwest. I have a lot of family over the range of this vehicle. However, for day to day operation, I wouldn't need more than 100 miles in range. A vehicle like this would be my car for driving to work and driving home.

For driving long distances, I'd just use a gas-based vehicle. I'd keep my old car around for that. Probably would need it half a dozen times during the year at most.

There will be some folks who will find the range of this car to be limiting. But how many compared to those who won't?

I think the real issue here is the calendar life of the battery. How long until I have to replace the batteries? That's what makes it easier for me to compare, apples to apples, the cost of the car vs. using gas.

Based on the info I read here: http://www.teslamotors.com/blog2/?p=39 .... I'm not too impressed with the thought of 50,000 life on the batteries.
Of course, hopefully whatever the battery replacement would be in 5 years of advancing battery technology would make the replacement battery a smart replacement for other reasons ...

Pardon me, but isn't the article about AES-256? So this is a much more significant drop in the number of bits.

Of course, I've only read the summary. This is slashdot, natch.

... new here. ;-)

Plagiarism is a symptom of professors only being involved in the last step: reviewing the final product.

Require the students to submit multiple drafts. Meet with them for 15 minutes each and discuss their thought processes on the ongoing paper. You'll get better final products, teach people not to procrastinate, and smoke-out people who have no involvement in their "own work."

What, can't do that because you have 60 students in a class? Well, there's part of the problem too.

We're trying to find a technology solution to a problem with less student-teacher interaction. Typical!

I wasn't even talking about MySQL. There's a lot more to Sun's I.P. than that. Owning Java alone is worth 7.4 billion to IBM.

I suspect that there was something other than the price that made IBM turn away; some other condition of the deal. If so, I wonder what the Sun board put at a higher value than the shareholder's return in the sale.

This was an intellectual property firesale. IBM = idiots. Congratulations to anyone who realized Sun stock was ridiculously undervalued; you deserve the profit you made by buying low.

Oracle suddenly has a great operating system, great server hardware, a popular database, and the de facto language of server-side business logic (other than COBOL.)

And IBM has built so much of its business on Java.

IBM should have just opened the piggy bank and it would have saved itself the world of hurt it now has in store.