I would think they have been getting royalties on episode reruns, since they were tracked down via a cue sheet, which has their names on it. Cue sheet is used by performing rights organizations to pay out royalties to performers. Given the story of the track's composition ("we need a song, you have four hours") my guess is that these folks have produced a large number of songs over the past 30 years that earn relatively small royalties per track. Add on top of that there are multiple tracks called "X-Files" just in this episode and I'd further guess there are plenty of songs that the musicians are getting paid for that they have forgotten over the years!
> I still mention maintaining this site on my CV and LinkedIn - disappointingly I've never been asked about it in an interview. I suspect most of the people doing the interviewing these days are too young to remember it.
This is astonishing to me. I check back to see if this site is still up once every year or two just to have a smile. If you were sitting across from me in an interview I am quite sure I'd lose all pretense of professionalism and ask you about nothing else for the hour.
It isn't quite dead yet. It is just entirely Flash based so it likely won't make it to 2021. Safari has already blocked Flash completely while the other major browsers require you to manually enable it. Odds are those too will begin blocking Flash completely sometime in the next few months with Adobe set to end official support at the end of the year. So maybe add zombo.com to your Chrome whitelist and visit it one more time, because soon the infinite won't be possible.
The SEC's response: "Nester explained that before staff can work on an issue that involves a company, they have to sell any holdings of stock in that firm. As a result, he said, there shouldn't be any surprise that a sale would precede the announcement of an enforcement action."
I don't know about anybody else, but this seems pretty plausible to me.
That's extremely amusing, since the probability of a enforcement action conditional on an investigation being started is much higher than the prior probability of an enforcement action. And if an enforcement action occurs, the price is only going to go down. So this is practically an institutionalized version of the insider trading the article is talking about!
But then you have conflict of interest if you're asking SEC employees to investigate companies they hold stock in. The incentive would be to not investigate or impose milder penalties.
I would argue that the people regulating the stock market should not be allowed to own stocks or any other sector investment, like an ETF focused on the health care industry. THey should still be allowed to own index based investment vehicles like an SP500 ETF.
Hmm. Given that in order to retire in the US you must invest in the stock market that would be a big requirement. Don't get me wrong, I am all for solving this problem, but simply not allowing the SEC employees to participate in the market would be a pretty good way to ruin their retirement.
An alternative would be to instead provide pensions. I don't really see how that happening given the "no handouts at any cost" attitude of the majority party in the House.
This would make joining SEC a very expensive career move for a lot of people, especially if they only planned to be there for a year or two, and were heavily into former employers or other specific stocks based on their experience in the stock market.
They have to assign their holdings to a 3rd party financial manager/fund/401k/IRA/etc and recuse themselves from any active management of their own portfolio. Not sure if this is an industry-wide requirement, but I had to do this when I worked for a small bond fund some years ago. Zero trading my own account allowed, and required to disclose all personal holdings and activity to the compliance officer who reported to General Counsel.
Yeah, when I looked at the data in the article, the pattern wasn't that SEC employees were systematically selling stocks they were investigating. Rather, it was that they were always selling, and refused to buy stocks under investigation. This is consistent with the SEC's policies, and also makes a lot of sense.
I would chalk this up to a quirk of statistics rather than any deliberate malfeasance. (Although, I'd note that the SEC policy that employees sell any stocks of a company that they will be investigating is good for the employees, because those companies are disproportionately likely to suffer a stock price penalty in the near future. Basically, the policy is a way of turning material non-public information into a market advantage without requiring that any individual employee act on that. But this situation exists in a lot of other cases, eg. stock option grants should systematically beat the market because companies that are financially healthy are more likely to hire people, and that financial health usually translates into hiring before it translates into profits.)
Do you see any conflict of interest in an employee at the SEC and buying stock in every SP500 company matching their market cap, then selling them right before the SEC investigates them? If someone did that with the SP500 they could easily beat the index. I think this is the definition of malfeasance.
I think it's the definition of "unfair advantage", but I don't think it's the definition of "malfeasance". The individual SEC employees have no agency here: they are required by policy to sell the stock of any companies they are investigating.
Moreover, it's not clear if any other policy would better serve the public interest. Requiring or allowing them to not to sell that stock is an even worse conflict of interest; it would incentivize them to never find a company guilty, because then their own holdings would drop. Requiring that they never own stocks would be a prohibitive restriction that would turn away many people who are most qualified for the job. The blind trust idea might work, but adds a lot of overhead for employees, and also requires another level of enforcement to make sure employees are not leaking information to the trustee.
A lot of people are really uncomfortable with the idea that someone might have an unfair advantage in financial markets, but oftentimes this is quite unavoidable. It's also pretty small-potatoes compared to the advantages that professional financiers get by being primary dealers for the Fed, or market makers, or having relationships with Wall Street policy makers, or hobnobbing with company CEOs on the golf course.
So.... if you can monitor the sales of all SEC employees; then you could copy-sell, or buy up their shares en-mass and very rapidly sell them for a profit?
For what it's worth, that's exactly what the rule is for journalists at The New York Times. To quote the relevant section from the handbook:
"No staff member may own stock or have any other financial
interest in a company, enterprise or industry that figures or
is likely to figure in coverage that he or she provides, edits,
packages or supervises regularly. A book editor, for example,
may not invest in a publishing house, a health writer in a
pharmaceutical company or a Pentagon reporter in a mutual
fund specializing in defense stocks. For this purpose an
industry is defined broadly; for example, a reporter responsible
for any segment of media coverage may not own any media
stock. “Stock” should be read to include futures, options,
rights, and speculative debt, as well as “sector” mutual funds
(those focused on one industry)."
Thanks for your kind words. As somewhat of a novice myself, I found the lack of documented real-world use cases frustrating, so I'm super happy to see this kind of feedback! I hope this can help you down the road should you make a second attempt at Kalman filtering.
The likelihood itself is a single number, but the algorithm is maximizing the likelihood based on a vector of 993 parameters representing the quality of a seat. Each x,y coordinate represents a seat in the ballpark, and each seat is mapped to one of those 993 values. The heatmap evolves as the vector approaches maximum likelihood.
