All signs point to yes. Only a few years after I converted from manually reference counting (retain/release) to garbage collection in order to reduce code, it looks like Apple maybe now moving to automatic reference counting (ARC). The links in this blog post provide most of the available information.

Information is relatively scarce at this point and I’d like to see a guide on transitioning from GC to ARC.

Based on what information is out there, I think that this is probably a good way to go. I’ve wanted to move some of my frameworks to iOS, but refrained from doing so simply because I’d have to add all retain/release management back in. Not only would that be a pain, but it’s more code… and more code sucks. This could be a good long term solution. We’ll see.

While shopping for a car this last month, I started to realized that buying a used car didn’t look very attractive. For some reason prices were, really, really high — and, for the first time, we considered buying a new car instead.

The day after explaining my theory to a coworker as to why prices were high, NPR aired this short piece detailing the phenomenon.

ONARI GLINTON: When the economy was really bad, car companies didn’t produce as much. So, there are fewer 2008, 2009, 2010 used cars. Less supply, and on the demand side…

…and of course with the economy coming back, but people still wanting to save money, the demand side for used cars is high.

One month later, the NYTimes reports that prices are still climbing:

Prices for used cars have been on an unprecedented upswing, and analysts do not foresee a leveling off during summer. The conditions are upsetting the long-held assumption that used cars are always better deals than new ones.

But they also point to the other part of the supply problem,

The supply problem has deepened following the March earthquake in Japan, which has produced shipment delays of vehicles and parts and curbed production in the United States.

So it’s really a combination of a bunch of different factors, with the bottom line being that it’s a really bad time to buy a used car.

Check out this graph of our monthly grocery bill over the last decade:

I did a five month moving average to get rid of the noise. To interpret this graph, you need to know that the first few data points are for just me, before Julie and I moved in together. After that,

• Julie and I moved in together June 2002.
• October 2009 we moved to the Bronx.
• October 2010 we moved to the Seattle area.

Here are some of the obvious conclusions to draw.

1. Obviously my grocery bill doubled when we moved in together, but it was remarkably stable for the next five years.
2. Then, in 2007, our bills increased by more than 50%!  NOTHING changed in our lives during that time. So that just reflects the genuine large price increase in groceries.
3. New York City is very, very expensive.
4. I think it’s too early to tell how much cheaper the Seattle area is than NYC.

My categorization system isn’t exact, either. That is to say that what qualifies as “groceries” is actually store-based. So, anything from Safeway was categorized as groceries… which makes sense. But, for something like Rite Aid, which provided both household goods and groceries in NYC, there could be some categorization issues. Same thing with Fred Meyer after our move to Seattle. Suffice to say, this is only an estimate.

My wife and I went from two cars when we moved in together in 2002, down to one car when we moved closer to our jobs in 2004, down to zero cars when we moved to New York City in 2009. Unfortunately we’re back up to two cars now. This prompted me to go back and dig up an old email in which I detailed the cost of ownership of one of our cars that we owned for over six years. Here’s a slightly revised version of the November 2009 email.

We bought a Chevy Prizm in February 2003 with about 40k miles and sold it in October 2009 with about 103k miles. During the cars first 15 months it was used to commute, but after that, we exclusively biked to work, so it was only used for road trips and errands around town.

In total, over this time period of 6 years and 8 months we spent approximately $25,000 on the car, or roughly $10.50 per day!!! This works out to 40 cents per mile. We recouped $4000 by selling the car, or $1.60 per day.

The individual costs over that period break down like this:

• $8119.00 original loan
• $511.15 interest paid on the loan
• $5124.03 fuel (which is 8.1 cents per mile in fuel alone)
• $5495.10 insurance (initially quite high, but eventually dropped to only $400 per year even with comprehensive)
• $5387.78 service (regular fluid changes, tires x2, A/C, MAF, brakes, wipers, etc. )
• $569.00 airport parking
• $91.00 campus parking
• $138.00 title and registration fees

Insurance would vary a lot for different drivers, but I think our average of $68 per month over that period is typical. Our service fee seems quite high, but only includes “incidental” repairs of the AC and MAF that total about $1500 of that. The rest is just the usual maintenance (we made sure we had good (but cheap) tires and brakes, which I’m sure not every one does). Of course, had we used the car to commute to work every day, we’d have spent $800 more dollars in parking costs over that time period, not to mention more on gas, service, etc.

Bottom line is that after selling it, we still spent approximately $9 per day or $.40 per mile to own a car. Cars are expensive!

I should add that I have almost no doubt our cost of ownership is well below average.

Conversation with my two year old this morning:
Me: “Are you going to go to the moon?”
Tevian: “Yes, rocket ship, Tevian.”
He points to himself, then pauses…. and finally,
Tevian: “Car seat.”

We had just watched a few rocket launches, so I think it must have been clear by that point that one might want to be strapped in.

Remember the QuickTake 100? I sure do. That’s marked the beginning of Apple’s short lived venture into the digital camera business. Personally I think it was a really good idea for Apple to get out of that market as soon as it became clear that both the consumer and professional camera makers were going to get serious.

However, that said, it just occurred to me that Apple is actually successfully competing in this market again…. but this time with the iPhone.

The best predictor of your earning potential after college is not your SAT scores or even which college you attended, but the average SAT score at the most elite colleges that you applied for, even if you got rejected. This result comes from a study by Stacy Dale and Alan B. Krueger as described by Kevin Drum. What a cool result!

I absolutely love the idea of demand based pricing for parking, and it looks like San Francisco has officially kicked off their pilot program.

The basic assumption is this: if you’re driving a car in a city, you want to be able to pull over and grab a parking spot without hunting around. In a practical sense that probably means you want about one or two free parking spots every block (80% utilization). How do you do that? You price it right. If you have a block with no free parking spots, then obviously demand is far outstripping supply, so you need to up the price. Conversely, if you have a block with too many free parking spot, you need to drop the price.

The benefits to this are enormous. If you think about it long enough, you realize that having 80% utilization actually maximizes the amount of cars stopping at each block. The obviously corollary is that this is good for commerce. You see, if the price of parking is too high, and parking spaces are underutilized, then you’re obviously not getting as many car-based shoppers as you could. But, the reverse is also true: if you price too low, you’ve filled up all the spot and many car-based shoppers had to divert elsewhere. You lost customers.

Another, less obvious benefit, is that you actually reduce traffic. In cases where there isn’t enough available parking, drivers add time (and distance circling) to their trip in order to hunt for parking spaces. If you have a parking space immediately available, you reduce the driving time and distance of the average car, and therefore reduce traffic.

Since the idea is that the pricing is variable throughout the day, as well as in the long term, in some ways it hard to imagine how the idea could fail. What would would a failure even look like? By definition the pricing is set to always keep a few free parking spaces.

I think that my source for all of this is Donald Shoup, I read some stuff last year in detail, but I’ll have to dig up a link later.

I got a shiny new magic mouse for my birthday only to discover that it doesn’t really work with my early 2009 Mac Pro. The symptoms are a stuttering mouse that occasionally drops its connection. Not good.

I opened up the case and poked around a bit and discovered that Apple buries the bluetooth antenna deep inside the aluminum case, with no attempt make it external. My grandfather, a former electrical engineer, was watching this operation exclaiming “they can’t be that stupid!”

Anyway, here’s the definitive thread on the matter. I’m just going to order and external bluetooth module and be done with whole deal. Needless to say, it’s quite annoying, especially on such a high end machine.

Update 06/06/2011: Turns out things haven’t been working out with my external Bluetooth adapter. You see, it depends on getting power from the USB, so each time I set my computer to sleep, it drops the connection to mouse. This means that each time I wake my computer from sleep, I need to turn the mouse off and back on again. So, now I’m going do a hardware fix instead, something I obviously should have done the first time. Sigh.

With the release of Mac OS X 10.7 imminent, Quicken 2007 will be dead. Knowing that this day has been coming, I’ve been periodically checking out Mac financial software over the last few years, but have never been impressed with the alternatives. Sadly, as clunky as Quicken 2007 is, it has a very robust feature set that (apparently) can’t be easily replicated.

Here are some notes for current alternatives. I didn’t actually check out the online banking in any of these, I’m just taking them at their word. If (when) I do think I’ve settled on one, I’ll have to check out the online banking in more detail.

iBank 4
iBank has come a long ways over the last few versions and could replace Quicken, but not without some sacrifices. For the first time it actually successfully imported all twelve years of Quicken data without any troubles. All the different types of accounts (including investments) seem to work. The layout is good to very good.

The big problem with iBank is the lack of reporting and graphing. From my experimenting, it’s really, really limited. With Quicken I like to (for example) blow up my net worth graph, binned by month, to my full 30 inch screen and examine the details. With iBank, I can’t even seem to get more than twelve bins, nor does it expand beyond 8.5 x 11 inches. Why the heck would they try to format to printed pages? I thought we were using computers these days, no?

So urgh, I really, really want to love iBank, but given that there’s (essentially) no graphing or meaningful reporting, it makes it difficult. It will probably work for balancing my accounts, but that’s about it.

Money
Okay, I tried version 3 and that simply didn’t have enough of the basic features to consider. It was able to import all of my data, but not without fouling up in the investment accounts. Version 4 looks good, but is currently in beta and very unstable. I managed to get data into it (after much waiting) by importing to version 3 and then upgrading, but it said that my investments were more than four billion dollars in the red.

The graphs are definitely better than iBank, but I still couldn’t get them to do some of the basic bar graphs over time.

MoneyDance
Hands down the most flexible of the lot, but hands down the ugliest too. It actually does the graphs and charts that I want, but their aspect ratios and fonts are oddly distorted. The whole program is just one big UI mess. It’s so bad that I won’t support it with my money — I don’t want to encourage that kind of shoddy development. Perhaps it goes without saying, but it’s cross-platform.

SEE Finance
Thanks to Brian for pointing this out, this one might actually be the winner. Imports appear to all be successful and it does have the graphs and reporting features that I want. I have some more experimenting to do, but this seems to be the winner at this point.