Monthly Archives: July 2012

Should you be a manager?

I recently enjoyed a conversation with our 2012 eBay interns. We discussed careers, leadership, business, and engineering. Someone asked me about career path: should I follow the manager or individual contributor path? It’s a great question.

The answer is it depends on what you’re passionate about, and ultimately that’ll be key in determining whether you’re good at it. Here’s my litmus test for the manager career track:

  • Are you passionate about leading people? If not, don’t become a manager. If yes, you need to develop people management skills: from growing people and helping them succeed, to delivering tough messages and handling challenging personal circumstances. You’ll need to spend much of your time working with people
  • Is having impact through others rewarding to you? If yes, that means you feel reward when your team hits its goals, the people around you solve problems, and your employees work together as team. If not, you’re someone who highly values personally contributing ideas, solving problems, or creating output (such as writing code)

There’s no right or wrong answer, and it isn’t black or white. You can be a good manager who still contributes personally, but realize its more about others than you. You can be a great individual contributor who’s passionate about helping others succeed; indeed, that’s a prerequisite of a senior individual contributor. But at the core, management is about leading others and being accountable for a team, and succeeding or failing based on their contribution.

Explaining Baseball to a Cricketer

I grew up watching and playing cricket. I’ve since fallen in love with baseball. They’re similar games: baseball was born from cricket. They both have subtlety, subplots, the ebb and flow of a long game. The games often contain something you’ve never seen before. I couldn’t find an interesting explanation of baseball for those who understand cricket — though I offer no promises of wikipedia’s (dull) comprehensiveness. This’ll be the first of two posts.

Getting out

You’ve read this far, so you must know some basics. There’s nine innings, each team bats in each inning, three outs and you change teams. When you’re out 27 times, it’s all over (unless it’s a draw – more on that later).

You can get out lots of different ways, just like cricket. The common ones are being caught, struck out, forced out, and tagged out. There’s also obscure ones – both cricket and baseball share rules around how long you have before you’re out for not batting in a timely fashion.

Australia playing the Seattle Mariners in 2009 in Peoria, Arizona. Australia was warming up for the World Baseball Classic, and the Mariners had just begun their Spring Training for the long season ahead. Australia won 11-9. The first game for Ken Griffey Jr with the Mariners in his second, unsuccessful stint with the team.

You get a strike against you as a batter when you swing the bat and miss, the ball passes through the strike zone without you hitting it, or you foul it off by hitting it outside the field of play (behind the lines that are on each side of the field, often marked at the ends by large yellow foul poles). You strike out when you do any of these three times – except you can’t get a third strike on a foul. Striking out is analogous to being bowled.

You get a free pass walk to first base if you’re hit by the ball (rather different to cricket). You also get a walk if you receive four pitches (so-called balls) that aren’t in the strike zone, and that you didn’t swing at. The strike zone is subtle: in the way LBW is subtle. It’s roughly knee height to just below the writing on the jersey, and roughly as wide as the home plate that sits on the ground.

You’re forced out when you have to advance to the next base, but you don’t make it. You must advance when you hit the ball into play (it always tippity run), or when you’re on a base and there’s a batter behind you who’s advancing to the base you’re on. Force outs typically happen by a fielder standing on the base you’re headed to and having the ball in their possession. It’s just like a keeper knocking the bails off in a run out, except it’s a run out where you were forced to run.

You’re tagged out when you’re touched by a fielder who’s holding the ball and you’re not safely on a base. This usually happens when you’ve decided to run to the next base when you weren’t forced to. This is a bit subtle: as a spectator, you have to know whether the runner is being forced or has chosen to run, so you know what to expect the fielder to do to get the out. This tagging the runner thing is something you don’t see in cricket.

There’s nine players, but sixteen 12th men

There are nine guys on the field when you’re fielding. One’s the pitcher, one’s the catcher (a wicketkeeper-like guy), and there’s seven fielders (position players). I’ll explain later where they all stand.

You can take any guy off the field, and replace him with any one of sixteen “12th men”. The catch is the guy that comes on the field really does replace the guy who left – he’s out of the game, and the new guy bats or pitches or both. The guy who left can’t come back – he’s a spectator until the next game. (I think this bears resemblance to the “super sub” they’ve experimented with a few times in cricket.)

In practice, there really isn’t sixteen 12th men. There are twenty-five guys on the team (“the 25 man roster”), but probably five or six of them won’t play on any given day. They’re usually pitchers who’ve pitched recently and are having a rest day, or are going to pitch tomorrow. And there’s always a backup catcher or two, which the team won’t bring into the game because they’re keeping him in reserve in case of injury (a bit like the backup goalkeeper in soccer).

So, really, you’re probably looking at ten guys who could come into the game: probably five or six pitchers, and four or five batters. In practice, you might typically see anywhere between zero and six of them. I’ll explain more on that topic later.

There’s an exception: in September, you’re allowed to have a 40 man roster. This is to allow teams to try out young players, and give the senior players some rest time before the postseason.

Batting order

Since there are nine guys per team, nine guys bat. They’re listed in a batting order, much like cricket. One bats at a time, there’s no non-striker’s end.

A lineup card that shows the batting order. These are exchanged between the teams before play — just like in cricket.

Once all nine guys have batted, they return to the top of the order, and the first guy bats again. There’s no limit on the number of times you bat – you just go around and around the batting order until the game is over.

If you follow Major League Baseball, you’ll know there’s two Leagues: the American League (which has 14 teams) and the National League (which has 16 teams). Most of the time, teams from each league play only teams from that league. The major exception is the World Series – the final, championship games of the year, where the best team from the AL plays the best team from the NL to be crowned world champion. (There’s also this thing called interleague play, I’ll ignore that for now.)

Anyway, the point I’m getting to is this: the pitcher in an American League team doesn’t bat. So, instead of having nine players on the team, they actually have ten: there’s a guy called the “designated hitter” (or DH) who bats in place of the pitcher. Technically, the DH could bat for any other player, but it’s always the pitcher who doesn’t bat. The DH doesn’t field, he just sits around, and gets probably four or five chances to bat in the game.

In the National League, there’s no DH, and the pitchers bat. Pitchers bat about as well as Murali or Glenn McGrath. There are no all-rounders in baseball.

It’s odd that the two leagues have different rules. You’ll find baseball purists like talking about this. Try expressing a strong opinion on the topic.

As an aside, cricket has played around with a DH-like rule in domestic cricket competitions.

There’s a lot more baseball

Baseball purists are staggered that test cricket matches last for five days. That freaks them out. They’re used to games of baseball that are two or three or (wow, that was a long game) four hours.

I’m more freaked out by baseball: the Major League Baseball season has 162 games per team per year, and that’s not including the postseason (the finals). They play from April to September, and so most teams get maybe three or four days without a game each month. That’s a lot of baseball – there’s pretty much baseball on all day, every day – in fact, there’s 2430 games per season.

Seattle Mariners 2012 schedule. 162 games, always 81 at home and 81 away.

Because there’s so much baseball, you can get a ticket cheap. Only the Boston Red Sox routinely sell out their stadium. A cheap seat might be $6.

Failure is vastly more common

A few ducks in a row, and you’re probably going to get dropped from your cricket team. Not in baseball. Failure is expected, at least it’s ok for quite a while.

A very good batter in baseball does something useful about 35% of the time. The other 65% of the time, he’s out without doing anything useful – kind of like a duck in cricket. What’s something useful? Well, that could in the extreme be scoring runs – anywhere from one to four (the maximum you can score by hitting a home run when there’s three other guys on the three bases). At the other end, something useful is getting yourself to first base – so that the next guy can help you work on getting around to score a run.

Strangely, something useful can sometimes be getting out – in baseball, the ball isn’t dead when you’re out (unless it’s the third out, in which case the inning is over). If you hit the ball a long way, it’s caught, there’s a guy on second or third base, and he can run around to home, he’ll score a run.

Interestingly, while 35% success is very good, 20% success is very bad. This is a very fine line: succeed 1 in 3 times and you’re amazing, succeed 1 in 4 times and you are ok, and succeed 1 in 5 times and you’re a disaster.

The ball doesn’t bounce

I guess everyone knows the ball isn’t supposed to bounce in baseball. If it does, it’s referred to as a “pitch in the dirt”, which sort of means it was a useless yorker. Since there’s no stumps. there’s no point in trying to sneak the ball under the bat.

The ball not bouncing means that movement off the pitch isn’t a factor in baseball. Instead, baseball is about movement in the air and varying pitch speeds – which you also see in cricket. Movement in the air in baseball is imparted by spin and seams, which in turn is controlled by hand and finger position on the ball. Most good pitchers in baseball have at least three pitches that they throw – a fast ball (they’re pretty much all fast bowlers), and a couple of “off-speed pitches” (slower, deceptive pitches that move around or don’t look slow out of the hand).

Draws don’t happen

There are no draws or tied games. After nine innings, if the score is tied, the game goes into a tenth inning where both teams get a chance to bat. If it’s still tied, it’s time for the eleventh. A ten, eleven, or even twelve inning game is reasonably common. I was at a 15 inning game once, you won’t see that often. The longest professional game went 33 innings.

There’s no overs

Pitchers pitch until their manager decides to take him out of the game and replace him with another pitcher. Baseball has managers who’re in charge of team (rather than a head coach) – they really do make the decisions. You’ll see them come out and visit the pitcher, and let him know he’s done for the game.

A starting pitcher (the guy who starts the game) will typically pitch around 100 pitches, maybe up to 120 if he’s a strong, veteran pitcher. That’s between 17 and 20 overs in a spell! Of course, if he’s getting hit all over the place, anything’s possible – he might last as few as 20 or 30 pitches.

When a starting pitcher is replaced, he’ll typically be replaced with a guy who pitches harder and faster for fewer pitches – a so-called relief pitcher. You might see one to four relief pitchers in a game, depending on how many innings the starting pitcher pitched. Relief pitchers are quirky, and they often have specialty roles. For example, there’s often a guy who’s a left-hander who specializes in throwing at left-handers, and he may throw only one pitch in a game before being replaced (the call these guys LOOGYs, lefty one out guys).

A starting pitcher warms up for quite a while before the game, probably more than you’d typically see a bowler warm up. There’s really no looseners in baseball – pitchers pitch at their top speed from the first pitch. To start each inning, a pitcher gets precisely eight practice throws to their catcher before the first batter steps in to bat.

The catcher is closer

The catcher in baseball fields like a wicketkeeper standing up to a spin bowler. The difference is the pitcher is always a fast bowler, and he’s about 6 feet closer than a bowler in cricket. A typical pitcher throws between 88 and 100 miles per hour (as the ball leaves the hand). Only the fastest fast bowlers have ever hit 100 miles an hour.

The umpire (in black), catcher (in red), and batter (in blue) in close proximity. It’s normal for the umpire to lean on the catcher.

This means that the catcher has to know where the ball is going to go, because he hasn’t sufficient time to react to the pitch. If you watch carefully, you’ll see the catcher tell the pitcher what pitch he should throw before each pitch. He does this by wiggling his fingers between his legs. If the pitcher likes the idea, he’ll nod and pitch. If he doesn’t, he’ll shake his head and the catcher will try another idea. The signals tell the pitcher what kind of pitch to throw (speed and spin), and where to throw it (high or low, in or out).

I’ll share more thoughts on the (subtle?) differences between the games next time.

Click Here!

Click here — the myriad of buttons on the web that are trying to pull user clicks

When a user clicks on a web page, it’s a good sign. They’re engaged with your content, they’re using your site. Clicking is certainly better than abandonment: where a user doesn’t interact with the page.

I recommend that you track percent abandonment of each of your pages: it’s a simple measure of how you’re doing in engaging customers. The lower the abandonment, the better.

There are a few exceptions, where abandonment doesn’t necessarily imply a problem. For example, a page that tells the time, or otherwise answers a question, doesn’t require a click. Indeed, a click might be bad news – the answer wasn’t a good one.

The more clicks, the more interesting the link to users?

Not necessarily. Suppose you’ve got a page with many links in many places. All things being equal, expect to see many more clicks on elements that are higher on the page. For elements at the same vertical height, expect to see more clicks on elements that are on the left of the page (the opposite is true for languages that are right-to-left such as Hebrew). All up, if a link is in the top left of the page, it’ll get clicked much more. In general, expect an extreme value distribution of the clicks.

Since users scan left to right, top to bottom, you should organize a page so that the most important things for users are closer to the top left, and the less important things are closer to the bottom right. This is pretty intuitive – you expect a web site to have the most common, useful links in its header, and the less useful more obscure stuff in the footer.

Hmmm. So, which links on my page do users like?

If you want to understand the relative performance of links on a page, you could consider swapping them and comparing the number of clicks. For example, suppose you’ve got a header on a page with three elements “About”, “Jobs”, and “Help”. You could measure the number of clicks on these for a week. You could then swap the “Help” and “About” links, and measure for another week. Does “Help” get more clicks in week two than “About” did in week one? If yes, the second ordering is better; if no, stick with the original ordering.

You need to be careful what you compare. It’s pretty safe to compare “Help” and “About” text in a header between two experiments. But you’ll find that there’s text you can create that will get more clicks, regardless of whether it’s more useful. It’s a well-known industry fact that “top ten” lists on tabloid sites get way more clicks than other stories. Text such as “click here” gets more clicks. Images attract the eye to nearby text, so that it gets more clicks. If you move something that customers are used to, expect them to click it less. And so on.

Experiments at scale

In a large-scale web business (such as eBay), we generally don’t do experiments sequentially in time. Instead, we’ll show the first alternative to some fraction of users, and the second alternative to another fraction of users. We can then compare the two populations over the same time, which both speeds up experimentation cycles, and also reduces any effects of seasonality or other differences between experiments that aren’t carried out at the same time. (There’s some issues this creates – a topic for another time.)

A Click is a Vote

If a user clicks on a link, this tells you something about that link. If it gets more clicks than you’re expecting, it’s usually good news (more on this topic later). If it gets less clicks than you’re expecting, it’s typically bad news.

What’s not widely known is that a click on a link tells you something about the links that come before it. Specifically, if a user skips a link on a page and clicks on the next link, this tells you that then former link isn’t relevant to the user. It doesn’t tell you links below it are irrelevant to the user — users leave the page when they see the first thing that’s relevant. This is a well-known phenomena in search engines.

Good clicks and bad clicks

A click is a good, basic signal. As I said at the start, it’s generally better to get a click than to not get one.

But there are ways you can make a click an even more reliable signal of user happiness. One simple trick is to factor in how long the user dwelled on whatever it was that was clicked. If they click on a link, press the back button immediately, and return to the original page, it’s actually a sign of unhappiness. They didn’t find what they wanted. If you wanted a rule of thumb, I’d say any click that dwells less than 10 seconds is a bad sign. I’d say any click that dwells more than 30 seconds is a good sign. You could try counting clicks, bad clicks, and good clicks, and drawing your conclusions from there.

Please share this post using the buttons below (hopefully they’ll pull your clicks!). See you next week!

Rebooting: a trick to avoid bugs

We all know that rebooting the home computer, router, backup device, DVR, or iPhone often solves mystery problems. (Have you noticed how frequently you’re rebooting your once-was-reliable iPhone?)

This works in large, distributed systems too. If you’ve got buggy code, a memory leak, or a shaky operating system, rebooting machines in a large distributed system works too. I’ve seen this in practice: periodic, scheduled reboots of boxes to reduce memory use, reduce CPU load, or generally cause a return to a known state.

Indeed, I’ve seen plenty of problems that occur when this doesn’t happen. A system remains untouched for a while, and things go south. After a problem has occured, I’ve heard quite a few folks say “we hadn’t rolled out code to that pool for a while” or “that box wasn’t rebooted for a few months”. In many cases, the issue was the gentle creep of increasing CPU or memory use.

Perhaps it’s good practice to ensure boxes are rebooted periodically. It’s probably wise when the machines are out-of-sight and out-of-mind: those less critical, less monitored, sometimes unowned services. It’s perhaps not even a bad thing: one of the wonderful properties of web services is they don’t have to be perfect, since you’re in control and the software’s running on your choice of hardware (unless you’re on someone’s virtual machine in some opaque cloud).

You, me, and the comma

Writing requires precision. You need to be clear.

There are five “Elementary Rules of Usage” that relate to commas in the legendary Strunk and White. I talk about one in this post, I’ll come back to the others in future posts. I’ve reproduced the relevant page below.

Strunk and White’s Elementary Rules of Usage. Using commas in lists is discussed at the top of the page. Click on the image to see an enlarged version.

Suppose you want to write a list of three or more things. Put a comma before the last item in the list. Here’s some examples:

  • The American and Australian flags are red, white, and blue.
  • The choices of shirts are red, blue, black, and white.
  • The city was bursting with cars, trains, automobiles, trucks, bicycles, and motorcycles.

The key point is that there’s a comma before the last item in the list.

Why’s this important? If you omit the comma, there’s ambiguity. Take the second example: “The choices of shirts are red, blue, black, and white”. It’s clear that there are four choices of shirt colors. If you omit the final comma, we’d have the following: “The choices of shirts are red, blue, black and white”. Are there three choices? Is the last choice a black and white shirt? Or are there four choices? If there were indeed three choices, this would be correctly written as “The choices of shirts are red, blue, and black and white”.

There is one exception to this rule. That’s when it’s a list of people in a company name. In that case, it’s obvious there’s no ambiguity. Here’s some examples:

  • Togut, Segal & Segal LLP
  • Amper, Politziner & Mattia
  • Berry, Dunn, McNeil & Parker

Grab a copy of Strunk and White (my latest copy is this beautifully illustrated, hardcover edition). Read the first ten pages (and then decide whether to read the rest, or pop it on your bookshelf and get street cred from your colleagues).