First off, this is coming from the prosecutorial side, who has every incentive to make the case as understandable as possible.

Frankly, it's not surprising that the average person can't understand complex finance and thus complex financial litigation.

> Everytime I hear a finance person use the excuse of 'you wouldn't understand' to a complex, I can't help but correlate it to a developer who's obfuscated code tell another developer they wouldn't understand, without acknowledging that they obfuscated the code.

The difference is that a jury is not other finance people. I don't expect jurors to understand financial crimes any more than I expect a random member of the public to understand a complex computer bug.

It's been my experience that this is very generally true, and you don't need the "finance" qualifier: people hate having their roles demystified to the point that they often don't even try to improve their own understanding of it.

Even if the person isn't greedy, they'll resent the loss of status that comes with, "wait, all you do is add entries on the CMS interface?" or "all you do is make the risks less obvious to auditors?"

As Mal says, "50% of the human race is middlemen, and they don't take kindly to being eliminated."

We should make a up a spurious tale about an enormous stargoat eating the sun and send them on ark to colonize another planet.

Watch out, you don't want to die of a public telephone transmitted virus.

Even good, clear, code is inscrutable complexity to a lay-person. Finance is no different.

To me, it's more than a little ironic how often software professionals look at what finance professionals do and ask the equivalent of "why the hell does Twitter need all this code? It's just 140 character messages!"

a fifteen minute conversation with a layperson about a particular code base will do just fine. "You know how a car is simple to operate, right? But you understand it has an engine and many moving parts? Well.. Twitter is just like that, and there are actually lots of small engines making sure the characters get from your keyboard to other people's eyes."

> ... you just don't want me to because then it would be obvious you are just ripping people off to make more money for yourself / company.

Agree. If your car pollutes more than other cars and runs on the blood of innocent consumers, hiding that in the explanation of how a regular car works is not an excuse.

Anyone who says "You wouldn't understand" is hiding something. If you can't explain it to me, do you even know what the fuck you're talking about?

So a guy is charged with tax accounting fraud for basically not putting some number into some entry in a spreadsheet when some external event happened, and instead putting that number into a different entry in the spreadsheet. You can't just say "you know how cars are complicated? Well so are spreadsheets. That's why this guy is guilty." Your analogy has to capture some legally-relevant dynamic of the system you are trying to explain.

Imagine your case depends on proving that only a guilty person would have failed to shard a particular database. It's not enough to just explain at the highest possible level the concept of splitting up a database. You gotta get the jury to understand it enough to know why you do it, and why it would be so out of the norm to not do it in a particular situation that the person must have had a nefarious motive.

I am not sure you can explain a buffer overflow bug that easily to a layman. That's still a trivial bug for a developper. Most of the products involved in recent financial scandals are relatively simple to finance professionals. The problem was rather pretty much always a gross under estimation of the underlying risk combined with excessive leverage.

"Imagine if I told you to go to the file cabinet and remove the 8th folder in the Mason file. Well, it turns out that the Mason file only has five folders, and neither of us were careful to think about the boundaries. So you go and count off seven folders after the first Mason, but then end up grabbing a folder that's three into the Mitchell file.

"The data on a computer is organized into ordered bytes like the file cabinet, and telling a program to go to some byte's location plus some offset can result in it accessing data that it shouldn't touch, if other safeguards aren't in place."

And that lets you Hax0r the computer because?

That's a great analogy, by the way.

"Ok, so say I need you to go and put this note of death at the end of the Mason file; that would be the 20th folder. You go to the drawer, count up to 19 from the first Mason folder and put the note there. You just ended up declaring Mr. Moriarty dead. Mr. Łeldon here will be processing credit card applications later in the afternoon; I'm pretty sure that Mr. Moriarty isn't going to get his now.

This is the way one uses the 'buffer overflow' problem in computers to manipulate data. With a little skill you can use it to do serious damage. Consider, for instance, what would happen if we were at an engineering company and slipped in a new blueprint for a machine instead of a personell note. Later someone else would pick up the design and the factory, none the wiser, would proceed to build our machine, instead of the one they wanted.

I don't buy the "you wouldn't understand" either. Here are simple explanations of a buffer overflow:

- We agree I'll give you max 1L of water, so you prepare a 1L bucket. Then I pour 2L of water into your bucket - you end up with a mess. (exploit version: We agree that if any water hits the ground, I can force you to do anything.)

- I give you card with typical form to fill in (squares for every letter). First 5 squares are for first name, next 5 are for last name. You fill in "Michael" as your first name overflowing into last name space.

- etc...

Sure, it doesn't capture all intricacies of the mechanism, implementation, side effects of the overflow, etc. But don't tell anyone they wouldn't understand the concept.

> Twitter is just like that, and there are actually lots of small engines making sure the characters get from your keyboard to other people's eyes."

That's an abstract explanation of what Twitter's code is, it doesn't get you anywhere closer to being able to actually understand the code.

Just like I wouldn't expect a layperson person to be able to find and fix a bug in a codebase, I don't think they can reliably evaluate the nuances of the financial system.

Your analogies don't help at all when it comes time to understand the details of said system, and evaluate whether or not a correct decision was made. When you need to reason about the details about a complex system, all analogies fall apart. The juries need to reason about details in a complex system and comparing Twitter to a "car engine" won't help a lay person understand why Twitter's fictional database layer is deadlocking and why the repair function shouldn't run during a transaction.

>If your car pollutes more than other cars and runs on the blood of innocent consumers, hiding that in the explanation of how a regular car works is not an excuse.

I think the takeaway here, is in the real world cars don't visibly run on the blood of innocent consumers. The means of procurement of gasoline is wrapped up in many different corporations, industries and nation-states which makes it incredibly difficult to ascertain blame.

I could make the argument that your volkswagen runs on the blood of innocent servants in Saudi Arabia. Who is at fault? Mercedes, Chevron, The Royal House of Saudi, A select few in that group, The American Government, or you?

I don't know about you particularly but I'm certain that there are a great many people who think that the financial system is ripping them off somehow when they really don't understand what's going on. Just look at High Frequency Trading & the Michael Lewis book. It comes up on HN every month or three and there are always people convinced that somehow these traders are "jumping in front of" other people's orders to steal their money when this is definitely not the case.

This sort of plausible deniability depends on there being financial instruments which are genuinely difficult to understand. I'm sure you would agree that there are in fact mechanisms and systems in computer programming which would be fairly difficult to explain quickly. Especially in terms which are anything approaching accuracy.

As software professionals we (some of us take) pride in trying to build stable systems that are decoupled and not overly complex. Financial systems seem to be build the opposite way.

Yet think of surviorship bias, isn't the code most likely to end up in front of a jury the hackiest, most complicated, least understandable, least documented stuff out there?

Think of the VW "scandal" at what point does a piece of code go from being a really effective emissions blocker to being an illegal emissions blocker? If they implemented and trained up a neural net then what piece of code is the violator, the guy who wrote the unit tests the neural net trained on, or the evolved coefficients in the network that humans never wrote, or the manager who told them to write an emissions routine using a NN or the regulators who entrapped them by knowingly writing bad regs with the intention of the .gov selecting winners and losers by selective prosecution?

Richard Bookstaber makes very similar points about the financial system in 'A Demon of Our Own Design', highly recommended.

When the investment banks were partnerships, the partners had plenty of 'skin in the game', and favored stable systems. Going public took the pressure off of the leadership's individual fortunes... and stability seems to have suffered.

They only seem that way to you because you do not deal in finance. A system can only be as simple as the business needs it's supporting whether than be a piece of code or a financial product. Some pieces of code are VERY complex because the business demands it. Same goes for financial instruments.

Is there a real reason for them to be so complex though? "Buisness demands it" isn't really a reason for me. It clearly affects the stability of the system, so shouldn't we look at ways of improving it overall?

Well you can't just say "them." That would be like saying "is there a reason that all that software is so complicated?" There are lots and lots and lots of different reasons for all the different pieces of code in the world. Same thing for financial products.

A finance person would probably say the exact opposite - most deals are just stacking two or three obvious, properly factored/decoupled abstractions together, whereas software people seem (to a finance person) to deliberately complicate things.

Keep in mind that in this circumstance, it's not the finance people who are claiming complexity. It appears to be the reaction from the jurors based on the case made by the prosecutors.

I wonder if a form of activism here might be to seek out Jury duty as a way to make sure the Jury does understand. To be honest I have never even thought about walking down to the Superior Court building and saying "Hi, I'm here to volunteer to be on a jury, what do you have today?"

This raises an interesting point, but neglects the fact that it's not possible (for good reason if you think about it) to simply volunteer for jury duty. In order to be eligible in most (all?) jurisdictions one must simply register to vote.

Neglecting the fallacy of the presumption that one can just volunteer, it is interesting to think about the way the world might be shaped should "more educated" individuals be the ones who solely shape public policy and case law. It's not very democratic in the pure sense of the meaning, but certainly interesting as a thought experiment.

>In order to be eligible in most (all?) jurisdictions one must simply register to vote.

I've seen states at least claim that being registered to vote is not necessary to be called for jury duty. It gets a bit hand-wavy after that but supposedly they also draw from, at least, driver licenses as well.

Simply filing state income taxes as a resident is enough to get you in the (potential) jury pool for your locality, at least in NY.

In my experience appearing too excited about performing jury duty is a sure way to not be selected.

Check out the book "The Bankers' New Clothes," which addresses the intentionally obscure language of finance