It was quite interesting to watch Jeopardy! this morning to see Watson take on Brad Rutter and Ken Jennings again.  While it is notable to see that had this been a one-day game Watson would have won, there were some particularly interesting points in the game.

The first was the Double Jeopardy in which Watson didn't know the answer/question but still had to provide one.  It guessed.  Think about that for a second: a guess is really just a set of solutions we're not sure about but propose anyway.  This is exactly what Watson did - he had a set of answers and picked the one with the highest confidence level (still below an actual confidence level), which as it turned out was correct.  So in that sense, we just witnessed the dawn of computers guessing.  That's fairly telling, because that's one more thing that computers can now do that before was purely in the purview of humanity.

Such guessing was more telling in Final Jeopardy!, when the category was US Cities and Watson guessed Toronto instead of Chicago.  I'm not quite sure whether it was over-emphasis on Natural Language Processing (NLP) on the part of the IBM researchers, but there should have been a check somewhere in the code that looked at the category title and combined it with the possible answers.  I'm left wondering if Watson's performance has looked solely at the questions this entire time without taking into account the boundaries based on categories.

I've also been impressed with Alex Trebek's handling of the whole thing.  He treats Watson as any other contestant, which is exemplary, though he didn't attempt to shake Watson's hand.  Of course he wouldn't, as Watson has no hands!  However, it made me think that in the not too distant future that there may come a Jeopardy! show where Trebek will end up having to shake the hand of the first robot to play.

Finally, is it me or did Brad Rutter look quite annoyed at the end of the game?  If I had to guess (ha), I would suggest that it's not because he was beaten, but because he knew all the answers!  I have a feeling that he just can't signal fast enough, as he only managed to buzz in on a couple of questions, and in those questions Watson wasn't buzzing in anyway.  Ken Jennings doesn't appear to be signaling in much at all, and most of his answers came when he had control of the board.

It'll be interesting to see what happens tonight!

So as the follow up to the story about my house being broken into (http://thomasquinlan.com/paying-taxes-is-supposed-to-get-us-something) I got my mortgage statement today from Bank of America.  My mortgage payment is going up $200 a month because my escrow balance has apparently gone down to the point where they have to start collecting more money.  This is a projection they've sent me that will start in April.  As it turns out, I do pay city taxes separately, and in addition to school taxes - the school taxes for 2010 were just under $1740 for the year.  The city taxes were $2045 for the year.  So of the taxes I paid on my house in 2010, 54% of them went to the City of Binghamton, for which I got nothing in return.  (I still don't even have a copy of the police report!)

The projection for the taxes for this year is interesting: $1740 for school taxes, and $2770 for the City of Binghamton for the year.  Yes, that's right, my City of Binghamton taxes are going up $725 dollars, or an increase of about 35%.  So, while the efforts of the City at providing youngsters with quality education will remain unchanged, efforts at police vigilance and other fine City services that I don't get will cost me 35% more.

So I've had an account with Morgan Stanley for a few years now, due to IPO stock that was purchased at a prior employer through a "Friends and Family" program.  It was 150 shares of stock, and the company in which the shares were held has had its valuation drop significantly (as did many others) in the Second Depression we're all in.

Yesterday, Morgan Stanley sold 119 of the shares of stock without telling me, to pay fees on the account.  I admit, I didn't read the statements, because if I had, I would never have agreed to keep an account charging me $25/month on what was at best $1750 worth of stock.  However, what I object to is not necessarily the fees, but the fact that they sold the stock without telling me.

I'm sure somewhere in the fine print it said they could do that when the account was set up.  However, it seems to me that a normal business that has unpaid account fees would make some effort to contact the customer to have those debts resolved.  I might even have sold the stock myself to cover the account fees - but I would have liked to have had the CHOICE.  To make matters worse, because Morgan Stanley didn't have an updated W-9 (something else for which they might have contacted me!) the IRS is now going to withhold 28% of the money even though I have had it for more than a year and the stock was sold at a loss.  Now I have to go get my money back from the government because Morgan Stanley can't send emails from its system saying "We need an updated W-9".  (It's an easy form to fill out and send, and it would have taken me less than ten minutes if I had known it was needed!)

I've liquidated the account with them now, so I won't be doing business with them ever again.  As if I needed further convincing, they charged me $21.95 commission and a $6 'maintenance fee' to sell the other 31 shares of stock.  A $250 transaction cost more than 10% with them (to say nothing of the additional 28% of THAT money that I will now have to get from the IRS).  At Schwab, my commission is $9 with no 'maintenance fee', about a third of what Morgan Stanley charges.

Maybe I'm not the type of investor that Morgan Stanley wants as a client - it must have been really difficult for them to have let their automated systems care for the 150 shares I had.  Maybe I'm not rich enough for them.  To be honest, I probably wouldn't have ever had an account with them if it weren't required under the terms of the stock program, so the feeling is mutual.  What I can be sure of is that if I ever DO become rich enough to need a serious broker (and I'm working on it), then I can be damn sure that Morgan Stanley won't be getting my business.

Apparently the unemployment rate has fallen to 9%, which is supposedly a good thing.  If it were a meaningful statistic, it would be; however, it's a ridiculous number.  There are several reasons for this (which the media does report, which is good), not the least of which is the fact that when people stop looking for work they no longer get counted.

However, there are a couple of interesting things to note with the most recent story (which can be found here: http://www.forbes.com/feeds/ap/2011/02/04/general-financials-us-economy_8292705.html).

For starters, the government has begun to count the 'self-employed'.  I hope they have criteria for that, but I'm going to go out on a limb and say that those criteria, if not fairly flexible, are made to be fairly flexible.

Secondly, this quote is telling:

"And the participation rate, which is the percentage of the working-age population working or looking for work, fell to a 26-year low of 64.2 percent."

That means that of all the people of working age, less than 65% are either working or looking for work.  Even if we assume that 10 points of the percentage of that is from people going back to school or otherwise deciding not to work, that's still a full 25% unemployment rate.  (One would also assume that people going to school would still want to find work eventually.)  That's IF we assume those ten points are from that.  It's quite possible that the actual unemployment rate could be in the 25-35% range, which is a far cry from the 9% as reported.