A bit more on persuadables

Following up on the last post, persuadables are one of those ideas with a high utility to difficulty ratio. They aren't hard to understand but they come in handy. So I thought I'd tack on a simple but fairly realistic example.

You own a casual dining place, think Red Lobster. You're sending out good coupons -- a popular twenty dollar entree for  fifteen  (still leaving you a profit of about a few dollars) -- to a random selection of people in the area. You've seen an uptick in business associated with the offer and you're seeing new faces (always a good thing), but the mailings cost you a quarter a piece and with a response rate of about two percent, the campaign is costing you a lot of money.

So you hire a statistician who builds a logistic regression model that lets you rank recipients and you only mail people who are highly likely to respond. Your response rate is now ten percent. A while later, though, you notice that your number of customers since you started using the model is back down to pre-coupon levels and your profits are way down. What happened?

The explanation lies in the three primary kinds of people who got your coupon in the mail. The first group is non-responders. They cost you a quarter a piece. Next are people who normally wouldn't have gone to your restaurant but decided to because of the coupon. These are the ones you like; they bring in money and may go on to become regulars. Finally, there are people who used the coupon but would have come by even without it. Giving them a coupon represents five and a quarter in lost revenue.

Remember, the model was built to rank likelihood to respond and as a general rule, the people most likely to come by after receiving a coupon are the people who would have come by anyway. By mailing only to the top deciles, you effectively selected the worst possible customers to market to.

This may seem like an obvious mistake, but it's not that unusual. Most people  who've worked with marketing analytics can come up with a few examples, some of which came with hefty price tags.

Facebook edges closer to underwear gnome territory

Following up on

Loan Sharks, Facebook and the Growth Fetish

All Things Considered chronicles the continuing slide from innovative company with good business model to cautionary tale:

For years the vast majority of business stories about Facebook seemed to describe some kind of unstoppable force. A company that was doubling in size every year — with members who spent more time on the network month after month, and a user base so large it rivaled the most populous nations on Earth.

And then as Facebook prepared to go public, suddenly the only thing business writers and analysts were interested in talking about was the money.

Analyst Debra Aho Williamson says Facebook's projected annual revenues are now more than $1 billion less than what she and other analysts had expected before the company went public.

"So Facebook is now in the ranks of the big companies, and essentially it'll stop innovating the way we've seen innovation before," says technology researcher Vivek Wadhwa.

Anant Sundaram, with the Tuck School of Business at Dartmouth, says, "We're talking about Facebook having to grow its revenues at between 25 and 30 percent per year."

Facebook Stories

They are among many analysts who recently have expressed their skepticism about Facebook's future as a business. Now Facebook is trying to change this narrative. The company is launching Thursday what it calls a micro-site, facebookstories.com.

It's really an online monthly magazine devoted to stories on how Facebook can be used.

The story featured is a genuinely touching story of a man who lost his memory trying to rediscover his past on Facebook. From a business standpoint, though, the micro-site is a pointless exercise, and if the executives at FB actually expect this to move the needle, the company is in even more trouble than we thought.



The problem here is one that almost every business analyst has faced. You want to convince people to do something (buy your product, not cancel their membership, spend more time on your site, whatever). These people can be put into three groups, lost causes who will never do what you want, sure things and persuadables (people who will do what you want given the right pitch or incentive). This last group is why marketing departments exist (and why marketing statisticians  like me have jobs).

FB has to meet incredibly aggressive growth targets that require both bringing in more accounts (extraordinarily difficult for a product this close to saturation) and getting more activity from many millions of low traffic users. The stories site may have a slight positive effect on brand but for the problem at hand it's so comically inadequate that it raises questions about whether the people running the company realize how much trouble they're in.

For starters, Facebook is too big and too famous to get much of a PR bump out of, well, anything. These "extraordinary stories" would be of great value to a little known business but people already know about FB and how to use it.

Second, for this to make a difference it would have to change the behaviors of at least tens of millions of persuadables, but by its nature this type of site will be most likely to attract regular/heavy Facebook users (sure things).

Finally, not every persuadable will be persuaded by every message. Watching an inspiration video might well make someone more likely to spend more time on FB but how much more likely?

If you run any kind of realistic numbers through this scenario, you won't get more than a trivial impact, and when you consider that at the end of the month, employees are probably going to start dumping their shares en masse, initiatives with trivial impact shouldn't be on anyone's schedule now.

"I have recently seen the silliest film."

I just saw a film heavily influenced by Metropolis* and I don't mean that in a good way. Not that I have a low opinion of Lang's movie. It's an often stunning collection of stunning images and some remarkable action sequences in the climax, but it really is a silly story with fascist elements (Goebbels was a fan though not so much of Lang's later work).

H.G. Wells wrote a memorable if brutal review. You should read whole thing but for now here's a taste of the opening.

I have recently seen the silliest film.

I do not believe it would be possible to make one sillier.

And as this film sets out to display the way the world is going,

I think [my book] The Way the World is Going may very well concern itself with this film.

It is called Metropolis, it comes from the great Ufa studios in Germany, and the public is given to understand that it has been produced at enormous cost.

It gives in one eddying concentration almost every possible foolishness, cliché, platitude, and muddlement about mechanical progress and progress in general served up with a sauce of sentimentality that is all its own.

* If you think you know, you probably do.

Once again, intellectual property law has made satire superfluous

A couple of months ago I wrote a post about Apple patenting wedge shaped tablet computers. For a title I tried to think of a design feature that no one would possibly have the gall to patent. I came up with this:

Has anyone patented rounded corners yet? If not, I see an opportunity

Then this morning I hear the following on Marketplace:

Kai Ryssdal: A high-stakes and high-tech game of 'Did TOO! Did NOT!' got going today in federal court in San Jose. Apple is suing Samsung for $2.5 billion for -- Apple says -- stealing the designs for the iPad and iPhone. Samsung, in turn, says Apple stole its design from somebody else to begin with. 

As for the technology at the heart of this whole thing? Marketplace's Queena Kim explains that might not really matter at all.

---

Queena Kim: Paulette Taylor has been a jury consultant on patent lawsuits for about 20 years. She says for jurors when you’re called to a patent case: 

Paulette Taylor: It’s like being set down in the middle of foreign land where you don’t speak the language and you have no ideas what the customs are. 

And so, she says, the lawyer becomes the jury’s guide to this strange place. In the case of Apple versus Samsung, I’ll quote from Apple’s “pre-trial brief” here, the jury must decide whether Samsung violated “Apple’s D’677 and D’087 iPhone design patents and D’889 tablet design patent.” Translation: Did Samsung copy the rectangular shape, rounded corners and the button at the bottom of the iPhone.

Journalistic outsourcing

This American Life digs into the future of journalism and discovers that, not surprisingly, it uses a fake name:

And in fact in all the time he's worked at Journatic, he's never spoken directly to his supervising editor, who sits in St. Louis. They communicate exclusively through the computer. When Ryan has a question about how to do something, his supervisor sometimes answers by posting a private video on YouTube. That's the only time Ryan's ever heard the guy's voice. It's all very future.

After he got the student of the week assignment, Ryan was put in charge of editing death notices and little business briefs for Newsday, a newspaper on Long Island.

Ryan Smith And my job was just to copy edit it and again, put it in the database.

Sarah Koenig And where were these stories coming from? Like who was doing the writing that you were editing?

Ryan Smith I was told, actually, that they're located in the Philippines.

Sarah Koenig That everything you were editing was coming from the Philippines?

Ryan Smith That's what I was told at the time, that these obituaries and these business stories were all written in the Philippines.

Sarah Koenig I know, right? In the Philippines-- which means when you look up the death notice, say, of a man named Eugene Squeleney Jr. on newsday.com, you see a story about him. But there's no reporters name attached to it. It just says special to Newsday under the headline. But the story was actually written by someone in the Philippines. Journatics internal records listed her as Diana D. and say that Diane got the story from the obituary website legacy.com and just slightly rewrote the information there.

I called Newsday's newsroom to ask about this, why they would use Journatic for this work. I didn't get too far. Mostly I learned that Newsday has exciting hold music. I talked to an editor who didn't want to go on tape. When I pointed out the Squeleney story with no reporter's name on it, just special to Newsday, the editor said, quote, "I am totally unfamiliar with this. I don't know what it is," unquote.

Then I talked to a company spokesman who cheerfully said he'd look into it. But a few days later, I got an email saying he quote, "could not provide any information on this." He wouldn't even say whether Newsday was working with Journatic. Ryan says he learned about the Filipino writers after complaining to his supervisor that the copy he was getting was rife with basic grammar and spelling errors. That's when his editor told him to cut the writer some slack. They weren't native speakers. So Ryan wondered, why do we have these writers at all?

His editor wrote back quote, "well someone has to summarize the obits for the death briefs. And it's cheaper to pay an outsourced writer than to have an American writer editor do it. Unfortunately they're basically paid pennies for these. I have Filipinos asking for better pay on a regular basis. I wish I could do something for them." An ad Journatic placed seeking Filipino writers offered 0.35 to $0.40 per story. I confirmed with a Filipino writer that they are paid 35 to $0.40 a story and more for longer stuff. But wait, there's more. Here's Ryan.

Ryan Smith When I ended up looking at the names on a lot of the stories-- and the names on the stories that were published weren't the ones that I saw that had written the stories.

Sarah Koenig Here's what he's talking about. In the Chicago Tribune's local site covering the towns of Homewood and Flossmoor, for example, you can see that Eric and Joan So-and-so have listed for sale their 4 bedroom, 4.5 bath home on such-and-such a street for $695,000, that Eric is a general manager of a building company, that he attended Roosevelt University. And there's a picture of him.

The reporters name on the story is Jenny Cox. But there is no Jenny Cox. Or even if there is a Jenny Cox somewhere out there, she didn't write the story. The writer was someone named Giselle Bautista in the Philippines who works for Journatic. Again, looking at the computer system the company uses to manage its stories, it seems that when Giselle worked on this real estate story, there was a button called Select Alias. When she clicked on it, she had a choice. She could either be Jenny Cox or Glenda Smith.

Which lead to this gem of a defense from the CEO:

Brian Timpone Really what they're doing is assembling and copy editing a bunch of facts, right? So they write the lead. If there's a paragraph about a person, the paragraph is technically written by someone in the Philippines, but not written. It's like they have to type out who the person is, right? So they have to know how to write to send it over. I mean but to say oh, it's written in the Philippines-- I mean there might be a paragraph of it that the first draft is written in the Phillipines.

If you drop by the TAL site, make sure to checkout the fallout to the story.

What is in a name?

So I have been wondering for a while if the blog name, Observational Epidemiology, really reflects the direction of the blog.  It seems more like we have become a bit more broad; looking more at education, economics, the media and statistics (and less at my original focus).

Is it time to update this blog's title?

Red flags -- few quick thoughts on Hacker and algebra

Following up on the previous post, we do need to have a serious debate on curriculum, but most attempts at starting these discussions are undercut by poor understanding of the subtle and complex way that what we learn in the classroom prepares (or fails to prepare) us for, if you'll pardon the phrase, real life. This lack of understanding is often accompanied by warning signs and, with no disrespect intended toward Dr. Hacker, a number of these show up in this NYT op-ed.

Here are a few examples.

State regents and legislators — and much of the public — take it as self-evident that every young person should be made to master polynomial functions and parametric equations.

Using polynomial functions as an example makes me very, very nervous. Functions may be the most indispensable concept for real world applications and having a feel for them may well be most important foundation of mathematical intuition. This really bad example, particularly given his follow-up defense of "quantitative skills, critical for informed citizenship and personal finance." Has Hacker really thought through these skills?

Then there's this:

Nor will just passing grades suffice. Many colleges seek to raise their status by setting a high mathematics bar. Hence, they look for 700 on the math section of the SAT, a height attained in 2009 by only 9 percent of men and 4 percent of women. And it’s not just Ivy League colleges that do this: at schools like Vanderbilt, Rice and Washington University in St. Louis, applicants had best be legacies or athletes if they have scored less than 700 on their math SATs.

An old professor of mine, Bill Condon, once referred to the SAT as the toughest ninth grade math test you'll ever take. That about covers it. The questions are difficult but the algebra is extremely rudimentary and the geometry is so simple it can literally be covered in less than a page. The rest of the test is logic and basic  skills like reading graphs and tables. Nothing here is relevant to his proposal. Another really bad example.

I have to run now, but it's because I'm out of time, not out of red flags.

I need to read this in depth before I comment further

Though my initial impression is not positive.

From Andrew Hacker via Dana Goldstein:

A TYPICAL American school day finds some six million high school students and two million college freshmen struggling with algebra. In both high school and college, all too many students are expected to fail. Why do we subject American students to this ordeal? I’ve found myself moving toward the strong view that we shouldn’t.

My question extends beyond algebra and applies more broadly to the usual mathematics sequence, from geometry through calculus. State regents and legislators — and much of the public — take it as self-evident that every young person should be made to master polynomial functions and parametric equations.

There are many defenses of algebra and the virtue of learning it. Most of them sound reasonable on first hearing; many of them I once accepted. But the more I examine them, the clearer it seems that they are largely or wholly wrong — unsupported by research or evidence, or based on wishful logic. (I’m not talking about quantitative skills, critical for informed citizenship and personal finance, but a very different ballgame.)

This debate matters. Making mathematics mandatory prevents us from discovering and developing young talent. In the interest of maintaining rigor, we’re actually depleting our pool of brainpower. I say this as a writer and social scientist whose work relies heavily on the use of numbers. My aim is not to spare students from a difficult subject, but to call attention to the real problems we are causing by misdirecting precious resources.

Felix Salmon tries to keep'm honest

Not so long ago, journalism was a much more difficult job. Sources were only accessible through land lines and face to face conversations. Only brief documents could be transmitted. Most were unindexed; none were searchable. Misinformation did slip by from time to time but not that often, and when it did happen the response was an appropriate level of shame.

Now what would have been days worth of research can be done with a half hour of Starbucks WiFi and yet, if anything, documents are more likely to be skimmed and questionable statements from interested parties are more likely to go unchecked.

The following post from Felix Salmon shouldn't be exceptional -- all he does is take a close look at a widely covered report -- but we've reached the point where this level of journalistic professionalism is the exception.

Are crowdfunding statistics the new counterfeiting statistics? Certainly they seem to have become a meme. If you know that crowdfunding is a big deal, it’s probably because you read all about it in TechCrunch, in May (“these portals raised $1.5 billion and successfully funded more than 1 million campaigns in 2011″), USA Today, a few weeks later (“About $1.5 billion was raised in 2011 by about 450 crowd-sourcing Internet sites worldwide”), or maybe the Economist, a week after that (“$2.8 billion will be raised worldwide this year, up from $1.5 billion in 2011″). More recently, Forbes upped the ante even further: “This year alone, an estimated $3.2 billion dollars is expected to be raised through donation-based crowdfunding platforms like Kickstarter”.

All of these statistics, you won’t be surprised to hear, come from the same place: a May report from Crowdfunding.org and its research arm, Massolution. The report lists — by placing their logos on five successive pages of the report, so that their names can’t be searched — 135 different “participating companies”, starting with Lending Club and Kiva, and ending with… um, hang on a sec. Lending Club and Kiva? Since when are they “crowdfunding platforms”?

It turns out, if you look at the definition of a “crowdfunding platform” that the report uses, it’s incredibly broad: “an operator of a funding platform that facilitates monetary exchange between funders and fundraisers.” Which turns out to include not only peer-to-peer lenders but also FirstGiving, a website which non-profits use to accept donations, and which claims to have moved $1 billion of funds through its system. For that matter, the definition doesn’t even say that the crowdfunding platform needs to be online: I reckon that if anybody hosting a political fundraiser probably counts as a crowdfunding platform under this definition. Hell, the New York Stock Exchange would even qualify.

Oh, and guess what: if you add up all the money raised in 2011 from all 135 companies listed, it doesn’t come to $1.47 billion at all. It comes to just $575 million. Where does the other $895 million come from? The report basically pulls it out of thin air, reckoning that since it didn’t manage to get numbers from all of the crowdfunding companies in the world, it would try to extrapolate, somehow.

Loan Sharks, Facebook and the Growth Fetish

Think about the following game. I'm going to let you pick seven or twelve then I'm going to role a pair of dice. If you pick seven you win seven hundred dollars if seven turns up and you lose one hundred if it doesn't. If you pick twelve you win two thousand dollars if twelve turns up and lose one hundred if it doesn't.

I assume everyone who reads this blog would pick seven at this point but what if I added the following detail, there's a loan shark waiting outside with a large gun and he is going to shoot you dead if you don't hand him two thousand dollars.

Suddenly, that twelve is starting to look pretty good.

This example is extreme, of course, but it's not all that unrealistic. There are lots of situations where the value of money follows a nonlinear or even stair-step curve. (ever been a nickel short at the coke machine?). Unfortunately we often make all sorts of assumptions about smoothness and well-behaved functions when talking about finance and economics.

Think about stock options. Let's say you bought a share for one hundred dollars. Furthermore, lets say that the CEO (whose options are also pegged at one hundred) has a choice between two strategies: one has a fifty/fifty chance of a share price off 100 or 105; the other has basically a fifty/fifty chance of lowering the price to 50 or raising it to 110.

Which is the better strategy? That depends on whether we're talking about the point of view of you, the stockholder, or of the CEO.

This is, of course, another toy example, albeit a more realistic toy than the first. For a real and timely example consider this from Marketplace:

Facebook's ad revenue is growing steadily, but not fast enough to justify its stock price.

Williamson: Everybody is going to be looking for is some traction in mobile revenue.

If Facebook's ad sales aren't gaining momentum, the social network could face another problem: share overhang -- meaning, jittery employees could dump shares and the stock plummets. Since the IPO, employees haven't been allowed to sell shares. But starting in August, they will be.

Brian Wieser: If the revenue turns out to be way above expectations, it is possible to alleviate some of the share overhang.

Brian Wieser is an analyst at Pivotal Research Group. But he says if earnings disappoint, employees could stage an ugly sell off.

Other than the marketing analytics aspect (where Facebook is very sharp) I don't have any special expertise here, but to a casual observer, the company appears to be facing a seven/twelve choice.

The seven is to shore up the mobile side of the business, get costs under control,and focus on deepening the relationship with existing customers.This strategy has an excellent chance of paying off handsomely for many years.

The twelve is to try to try to grow fast enough to justify the stock price which is a hell of a long shot (When you're closing in on a billion members, a business plan that assumes further explosive growth may not be realistic). Unfortunately, even having a chance at that goal requires pouring money on the problem and squeezing more cash out of members in ever more annoying ways. This strategy runs the risk of leaving the company cash poor with a tarnished brand and unhappy members.

All of this takes us back to the growth fetish, the tendency to overvalue short term growth even when it may end up reducing the long term returns of a company.

Remember that ongoing conversation about the decline of journalism?

From Marketplace:

Vigeland: OK, excellent. Give us a rundown of some of the subjects that you have been quoted as an expert on. 

Holiday: Turntables, insomnia, barefoot running and -- my favorite, of course, is boat winterization. 

Vigeland: And do you own a boat? 

Holiday: I don't own a boat. I don't own a turntable either. And I don't suffer from insomnia. 

Vigeland: All right. So what was your ultimate goal here? Why bother with this kind of thing? Were you bored, you needed something to do? 

Holiday: No, absolutely not. So I discovered about a year ago, there's this service that's called HelpAReporterOut.com. Basically, the reporter needs a subject for a trend piece; I'm promoting something, I say, 'I'll be your trend subject,' and we trade. The reporter doesn't really have to do their job, and the source gets free publicity. So what I wanted to prove was that this backroom arrangement left the front door wide open to all sorts of media manipulation. And so, in about six months, I was able to deceive almost every major media outlet -- not a single one fact-checked, not a single one asked, 'Hey, why are you giving us this information? What's in it for you?' Look, sources have always been self-interested -- we've known this -- but we've almost embraced that self-interest and not asked: 'What are the consequences of doing so?'

More on microfoundations

Speaking of the stuff of the queue, Joseph and I have been talking about questionable defenses of economic rationality. Now Paul Krugman (who pumps this stuff out at an astounding rate) has another example of theory not matching reality:

So: I’m old enough (you kids get off my lawn) to remember the rise of self-service gas stations, which coincided with the oil shocks of the 1970s. Suddenly gas was much more expensive — and drivers, eager to limit the blow, were willing to pump their own to save a few pennies. 

At the time, being either in or fresh out of grad school, I thought, wait, this makes no sense: the decision to pump your own gas is about making a tradeoff between money and your own effort, and should have nothing special to do with the price of gasoline per se. Yet people acted as if they had a limited budget for fuel, as opposed to an overall budget constraint, and responded to a rise in gas prices with a seemingly irrational effort to hold down the size of that sub-budget. 

What the new paper by Hastings and Shapiro does is to show that the same kind of behavior continues to apply, with consumers shifting to lower-grade gasoline when gas prices rise, to higher-grade when they fall; the upshift happens even if incomes are falling. 

What does this have to do with the macro wars? Well, if the assumption of perfect rationality breaks down even in the most standard of micro settings — if consumers behave in a way inconsistent with full maximization even when doing something as mundane as choosing which type of gas to put in their tank — how absurd is it to insist that, say, Keynesian stories about the economy can’t be right because we can’t fully derive them from intertemporal maximization?

Two anecdotes on how (and how not) to run a business

I'm going to be discussing both of these businesses in future posts but since my queue is pretty full at the moment I thought I'd get these two examples out while they were still current.

The first involves Weigel Broadcasting, probably the best run business you've never heard of. As with sports and politics, there's an aesthetic pleasure to watching business done well and under that criteria, Weigel is in Joe Montana territory.

Take the response to the death of Andy Griffith on their MeTV network. The network ran a slate of shows featuring Griffith including the Make Room for Daddy back door pilot. Nothing particularly surprising there. I'm sure they plan these in advance and have already laid out the shows they'll air when other notables like Dick Van Dyke or Mary Tyler Moore pass away.

What was notable was the timing. The tribute aired on the Fourth of July. It was an inspired choice -- no living performer was more associated with Americana than Griffith -- but what makes it notable was the fact that Andy Griffith died on July the third.

Let's run through the timeline:

1. Decide on the Fourth

2. Reschedule the day's shows

3. Record the promos

4. Put the promos into heavy rotation

5. Issue press releases.

I've seen simpler corporate processes stretch on for months. At Weigel, this took six hours on the outside. If we had better business journalists, you'd be hearing more about Weigel.

Now for something completely different...

I was checking Hulu last night when I noticed an item about the Dark Knight. I immediately assumed it was something about the shootings (keep in mind, the time you see at the bottom of the screen is West Coast time) but instead it was a jokey piece on fake spoilers. It was still there when I went to bed.








When you get a big, tragic story like this, smart nimble businesses immediately ask themselves if there's a negative PR aspect that they need look out for and if possible, avoid. This is particularly true for websites because

1. it's easy to make changes

2. screen captures are forever.

I suspect that someone at Hulu saw this and thought "we really ought to pull that" but the company wasn't set up for that kind of rapid response. This is also consistent with other things we've seen from Hulu, but that's a topic for other posts.

(also posted at MippyvilleTV)

Krugman on Climate Change

Paul Krugman makes a couple of essential points about climate change. The first points to a new paper by James Hansen. As Krugman puts it:

The first is the relationship between extreme weather events and climate change. The normal, cautious thing is to say that there’s no way to attribute any particular event, like a heat wave in the Ukraine, to global warming — and news media have basically been bullied by this argument into rarely mentioning climate change even when reporting on extreme weather. But Hansen et al make an important point: this argument is much weaker when we’re talking about really extreme events, like temperatures more than 3 standard deviations above historical norms. Such events would almost never happen if there weren’t a rising trend in global temperatures; so when they become quite common, as they have, it’s fair to call them evidence of warming.

The second addresses something that has bothered me for a long time, the "won't there be winners with global warming?" argument. My rebuttal has always been that we have optimized agricultural land use based on our current climate. Krugman makes that point but takes it further:

The second point is how we know that climate change is a bad thing — a question I sometimes get asked. The questioners wonder why the fact that, say, more of Canada becomes agriculturally viable doesn’t offset the damage in places that get too hot. 

My first-pass answer is that we have a global economy that is adapted to historically normal climate — not just in terms of what is grown where, but in terms of where we locate our cities. In the long run, after a couple of centuries’ worth of urban development and infrastructure has been drowned by rising sea levels and/or made useless because previously habitable regions need to be abandoned, we might be able to reconstruct an equally productive economy; but in the long run …

On a related note, we are now having to worry about invasive species in in Antarctica.

Saturday Afternoon in LA

I'm typing this sitting under the umbrellas at Irv's Burgers, a Fifties era hamburger stand in West Hollywood. It's a notoriously friendly neighborhood place and it reminds me of something that first struck me when I moved to LA from Atlanta: for all its trendiness, LA has a distinctly old fashioned attitude toward dining. In most of the country, if you drive down the street looking for a quick bite to eat, you will see the same places serving the same food.

It's true that national chains are eating away at local dining, but the independents and local chains like Tommy's and Zankou's still dominate much of the town. Part of the reason is certainly the loyalty these spots inspire. Pulitzer Prize winning food writer Jonathan Gold once refused to say whether he preferred Apple Pan or Pie N' Burger because he didn't want to deal with the letters from the patrons of the spot he didn't pick.

I was having a cheeseburger at Apple Pan last week and I got into a conversation with the man sitting next to me. He had been coming there weekly for over fifty years, only slightly longer than the counterman had been there.

There are plenty of regulars here at Irv's, walking in and picking up old conversations through the pick-up window with the family that owns the place. There's a comfortable, small town vibe here that you can find in most of LA.  That's not something most of the world associates with LA, but the locals know.