The Subway Principle

At Rockerbox we're big believers in the Subway Principle.

We defined the principle in our company wiki as:

Subway Principle: If we get an email, but don't want to / can't respond right away with an answer, we respond right away acknowledging receipt of the email and indicating that a response will come. Better for them to know we saw the email than to hang in thin air.

This mostly comes into play when we get emails from clients, prospects or job applicants that we've interviewed (not for cold emails from vendors). The idea being that there's nothing worse for the sender than not knowing if an email was actually received and not having any expectation of when an answer will arrive.

Why is this called the Subway Principle? It stems from an article I read a while back that I'm unable to find (If you know the article I'm referencing please send it my way because I take zero credit for inventing this principle).

The premise of the article was about how you feel when you're waiting on the platform for your subway. If you're waiting with no indication of when the next subway will arrive, you'll get frustrated. You'll get frustrated even though the next subway may be arriving in 2 minutes. Simply not knowing is what causes the frustration.

On the other hand, you're more calm if you're waiting at a platform that has screens displaying expected arrival times of the next subway. Sure you'll be upset if you arrive to a 30 minute wait. But once you get over the initial anger, rather than just sitting around and twiddling your thumbs, you're in a position to figure out how to spend the 30 minutes.

With subways, it's better to know so you can plan than to be unsure. The same is true with customer emails.


Also published on Medium

The Cell Phone I want

Every-time a new iPhone comes out I instantly want it. Even if there are only superficial changes, I want the latest model. I've been a lover of technology my whole life so this isn't really anything new. The earliest memory I can recall of this was desperately wanting a laser printer when I was younger. At the time I was probably 10 years old or so. Why on earth I needed a laser printer, let alone any printer, is beyond me. Yet the moment I saw the first laser printer I wanted it. I was just so intrigued my a non ink based printer.

This brings me today and my desire for the newest iPhone. The ironic thing is that I fully recognize that my iPhone has by and large made me a worse person. I'm not unique or special in this regard. I'm as addicted to my phone as everyone else. It's made me less social, less able to handle 60 seconds of boredom and less able to concentrate on anything that requires more than a couple minutes of continual focus.

Just this morning, I left my house with my Kindle to get some coffee. I read in the coffee shop for ~60 minutes without barely a moment of distraction. It was just continual in-flow reading. I went home afterwards with the intention to keep reading. Lying on my couch with my Kindle, it wasn't even 2 minutes before I had a mental itch to check my cell phone. I noticed the desire and really made an effort to focus on reading. 3 minutes later, the itch returned and I unfortunately caved.

Now, I'm not a luddite by any means. I love technology and what is enables. "Any sufficiently advanced technology is indistinguishable from magic" is one of my favorite quotes. At the same time, I believe it's time to build technology that doesn't just enable us to do more but also enables us to live better lives.

So, in light of all of the above, what do I actually want in a cell phone? I think there are two ways to approach this. Each one is probably better for a different type of person:

1) Limitation of functionality: I'd essentially be looking for functionality that lets me keep in touch with friends and family, get around, take photos and learn. Specific apps would be the phone app, texting (this can include apps like WhatsApp and Facebook Messenger), maps, camera and Kindle. This phone wouldn't have any notifications except for incoming phone calls from people that I know.

2) Limitation of time: In this case the phone would limit the amount of time that people can spend in certain applications. On some cadence (weekly, monthly?) I'd let my phone know how much time I want to spend per day on things like Twitter, social media, browsing the web and others. Once that limit is reached for the day my phone would automatically terminate my ability to use these apps until tomorrow.

Personally, I think the limitation of functionality would be better for me but I recognize that it's a bit more of a baptism by fire than the limitation of time approach.

I think Apple is in a better position to enable both of these options than Google. Apple's business model isn't built around advertising but rather making products that people love. I think a product that makes me feel better by forcing me to be more cognizant of how I spend my time would in fact be a better end product. I don't really see the downside for Apple in enabling this.

Google needs people to be browsing the web and consuming for it's advertising business model to work. I don't really see them enabling this. On the other hand, Android being a more open platform could let developers toadd on this functionality faster than Apple's more closed operating system.

So where does this leave me? Am I going to by the iPhone X? You bet. But here's to hoping for theday where it comes with a more limited App Store.


Also published on Medium

Israeli Palestinian Game Theory

The Israeli-Palestinian conflict has gotten even more news coverage in the last few months than usual. From new resolutions at the United Nations to public feuding between Obama and Netanyahu to a French-led Mideast summit in Paris, the conflict seems to be top of mind for the world again.

Most of the attempts to solve the conflict to date have focused on getting Israel and Palestinians to directly negotiate (Madrid in 91, Oslo in 93 and Camp David in 2000). Other attempts include Arab led peace proposals (Beirut Summit in 2002), United Nations led solutions (starting with the original UN partition plan through resolution 242) and international summits like we’re seeing now in Paris.

The point of this post isn’t to debate which approach is best. Instead, it’s to outline a game theory based approach that’s been in the back of my mind since I took a course on game theory at McGill.


For people who don’t know what game theory is, check out wikipedia. It’s been so long since I took the course that I don’t even know if the below formally counts as game theory. Nevertheless, I like how it sounds so we’ll give it a go.


Premise: Israel will unilaterally develop a set rules. These rules will outline how Palestinian actions on the ground will result in changes to the status quo. Lack of the action (as defined by Israel) will improve the situation for the Palestinians. Occurrence of the actions will worsen the situation. These rules will be made public (literally on a website) and tracked for everyone to see.


How could this work in practice? Here is an example of one rule:

Rule Regarding Impact of Terrorist Attacks

This rule outlines the way in which terrorist attacks impact the expansion or contraction of settlements.

Action: Terrorist attacks occurring within a given month.

Lack of Action: If no terrorist attacks occur within a given month, Israel pauses construction within settlements in that given month.

Occurrence of Action : If a terrorist attack occurs within a given month, Israel will continue construction within settlements in that given month.

Lack of Action Streak: For each subsequent month with a lack of action (i.e. X months in a row with no attacks), Israel will pause construction of new settlements.

Occurrence of Action Streak: For each subsequent month with an occurrence of action (i.e. X months in a row with an attack), Israel will build a new settlement.

Lack of Action Super-Streak: Every Y months in a row (say 6 months), Israel will evacuate residents from and destroy 1 settlement. For each consecutive Y month streak, the number of settlements evacuated and destroyed increases in a fibonacci pattern (i.e. 2 settlements after 12 months, 3 after 18, 5 after 25 etc…).

Occurrence of Action Super-Streak: Every Y months in a row (say 6 months), Israel will annex 1 settlement into Israel proper. For each consecutive Y month streak, the number of settlements annexed increases in a fibonacci pattern (i.e. 2 settlements after 12 months, 3 after 18, 5 after 25 etc…).


The above is an example of one rule. I can imagine lots of other rules being developed as well, for instance tying rocket attacks with prisoner releases, stone throwing with checkpoints etc…

The overarching goal is to put control in the hands of the Palestinians. If they want a state, the set of rules should be designed to incentive actions that will lead towards their having a state. At the same time, it should incentive results for Israel that promote security.

Making all of this public (i.e. the rules and how specific terrorist attacks adversely impacted them) will hopefully cause positive public pressure within the Palestinians. The idea here is that we want to change public opinion from being pro attacks to being upset at attacks as they’re directly causing a known and clearly pre-determined negative impact on the Palestinians.


I realize that coming up with a set of rules on Medium is extremely different than making this work in the real world with real people. I also realize that even if the above were to be practical, it won’t directly result in peace. Instead, I’m hoping that it’ll leads to conditions on the ground that make peace more realistic down the line.

I’ll caveat all of the above by saying that the concept is far from fully fleshed out. I’m posting this mostly to get my thoughts on paper and to get feedback from others. What about the above is good? What about it is stupid? How realistic or unrealistic is this? Please leave comments and let me know.

48 Hours with the Amazon Echo Dot

Over Thanksgiving weekend I bought both the Echo Dot and Google Home. They were both scheduled to arrive this Monday but the Google Home is yet to show up (thanks Fedex). I've had the Echo for 48 hours now and here are my initial thoughts.

The Good:

It's fun. I was in bed last night and I wanted to know the time (I don't have a clock besides my phone). Being able to say "Alexa, what time is it?" while lying in bed with my face in the pillow put a smile on my face. Plus, the answer was correct.

It feels new. Using the echo reminds me of the first time I used Snapchat. I wasn't fully sure what the point of Snapchat was so I had to guess, experiment, screw up and eventually figure it out. The Echo feels the same way. I don't know what the best way to interact with it is. I'm trying different things, seeing what works and enjoying the process.

The device is nice. So far the hardware seems perfectly suited for the job at hand. It's simple and does the job. Plus the lights on top are fun. They go off whenever Alexa speaks giving it a slight sense of humanity.

Using Alexa to play music works well. Being able to just say "Alexa, play the Beatles" is pretty awesome.

Alexa responds quickly. I was expecting a decent lag between asking a question and hearing a response. I'm pleasantly surprised at how fast Alexa is in answering. There's still room for improvement but pretty good start.

Being powered all the time works. It's subtle but it changes completely the interaction model. While I was getting dressed this morning I was able to ask "what's the weather" without wondering where Alexa is, if it's on or what's in my hands.

The Maybe:

Still skeptical of ordering by voice. This isn't super rational as I buy almost everything on Amazon so why wouldn't I be ok with ordering by voice? I'm not sure why but I'm still not. Maybe this will change with time but I still have the "I need to check out prices / reviews" before I buy mindset which isn't geared too well for voice.

The So-So (nothing is really that bad):

Alexa needs to get smarter. At the end of the day Amazon isn't Google yet. The answers Alexa gives aren't as good as a Google search. TBD on how good Google Home is there, but there's definitely a host of knowledge that Alexa needs to acquire. This includes questions like what's on TV tonight and how many points did Russell Westbrook score last night?

Alexa doesn't understand conversation sequence. Alexa is good at answering simple questions like "where was Tolstoy born." Following up with "when was he born?" gives me nothing. Alexa doesn't understand continuity of questions yet, a problem that's been pretty well reported to date.

Repeating "Alexa" within the same conversation is bad UX. Alexa should better handle conversations. I get a quick response when I ask "Alexa, what time is my first meeting tomorrow". When I follow up to the response it's annoying to have to say Alexa again, especially if my response is within ~2 seconds of the answer. I get that there're privacy concerns and that it'll get complicated if my next sentence is to someone else in the room. Still, this is something Amazon needs to tackle.

Third party integrations (skills) aren't great. When you use them you can tell they were developed by developers outside of Amazon. I tried the 7 minute workout skill. Starting a workout worked great. I spent the next 10 minutes trying to figure out how to cancel the workout midway (Alexa Exit, Alexa Home didn't work).

Packaging and setup need work. I got Alexa up and running quickly but I know Apple would've made the setup process 10x simpler and the packing 10x nicer.

Bottom Line: I'm really happy and excited by the Echo. It has a ton of potential and I'm excited to see how it continues to become a part of my daily life. I think it actually worked out for the better that I didn't get the Google Home and the Echo on the same day as I'll be in a better position to compare each one now.

Who wins? Kudos to Amazon for coming out with this. I think Apple could've released the same product at the same time as Echo with a pretty comparable user experience. Big stumble by Apple. My gut says this comes down more to Amazon vs Google than anything to do with Apple. The question is can Amazon become Google faster than Google can become Amazon? Specifically, can Amazon get its search results to Google's level before Google finds a non advertising way to monetize voice search outside of e-commerce?

Review of Do the Work by Steven Pressfield

Short and quick book to read. It's written in an interesting way. Very much the author speaking directly to you. Also uses sporadic huge fonts and random positions of text on page as a wa to emphasize points.

I enjoyed the book. It calls out the inherent challenge facing anyone that tries to do something. He calls this challenge Resistance. Pressfield says Resistance is a natural force, as much so as gravity or time or anything else. Resistance is constantly there and continually trying to prevent us from accomplishing our goals.

In order to complete anything, Resistance must be overcome. It's a fact that everyone who tries to do anything must accomplish. Basically, rather than being upset at the fact that there is Resistance, the right approach is to recognize that it's normal, and take the steps to work through it. That's the real accomplishment -- getting past Resistance.

This Resistance comes at all times. To start a project obviously. But just as much to finish a project.