Ripple effects

Last Friday my iPhone's vibrate feature failed. I had a day or two of odd brrrraap buzzes, wheezy ailing things, and on Saturday, pfft! no more vibrate.

I went to Apple's Genius Bar on Sunday, fighting masses of bored tourists on Easter to get my phone inspected. The technician (genius?) took a quick look at my phone and decided that I had broken the external silence switch when I dropped it at some point. "There's your problem, right there," he said cheerily.

Before I had the chance to get defensive, he had opened a drawer and taken out a small white box. Out came a new iPhone--refurbished, I'm sure, but visually perfect--and within five minutes the genius (technician) had swapped SIM cards and activated the new phone. He took my phone--nine months old, dropped several times, with the scuff marks to prove it--and put it in the box with an explanatory label.

And that was it. "Here you go," he said, "you're all set." And I went home with a new phone in my pocket.

I tell this story not simply to add to the "cult of Mac" but to examine just why Apple has been so successful.

• Trust. The tech who met me listened to my request, quickly verified it, and moved onto solving the problem. No challenges, no curiosities, no wondering whether I had violated an arcane clause of my limited warranty. Heck, the tech even pointed out that I had dropped the phone--surely grounds for voiding my claim, and for which I had prepared an extensive explanation about timing, cause and effect, and so on. But it made no material difference to him.


• Ease. All I did to get my phone replaced was make an appointment, hand over the phone, and sign a form acknowledging my receipt of a new one. No other paperwork or, as noted above, difficult questions.


• Flow. The Genius Bar is, of course, free. I booked online, arrived late on Easter Sunday, and still got taken within minutes.


• Goodwill. The net effect of the above: I am a newly satisfied Apple customer, not only proud of my iPhone (proud! of a phone!) but delighted with my recent experience. I've spent the week telling people my story, which routinely elicits amazement and wonder: what other company is this easy to work with? This in turn continues Apple's amazing halo effect, which translates into ever stronger sales.

The message, to any company selling products: treat customers with respect and make life easy for them. Individual transactions may have a higher cost than a cost accountant may prefer. But the long-term impact is undeniable.

Triangles of Doom

Cheap. Fast. Good. Pick two. That's the classic project triangle. As best as I can tell its an immutable law of the universe. No matter how much you try, you can only control 2 points at any given time. At Ai (driven by Wertheimer) we incorporate these points into a statement of priority from the clients, called a Faceted Feature Analysis. (Link will take you to the full story). Different clients are sensitive to some points on the triangle more than others, so we move the project priorities around in order to accommodate them.

This triangle is a source of confusion, however. That's because there is (or at least seems to be) more than one triangle in play. In fact there are three, making a triangle of triangles.

The second triangle is the Faceted Feature Analysis (FFA) triangle.

The mapping between the first two triangles is mind-bending, but legit. What we're saying here is:

• If Cost is the most important factor to a client, then Business Value is paramount.
• If Time is the most important factor to a client, then Technical Ease (how easy it is to build) is paramount.
• If Quality is the most important factor, then User Value (or usability) is paramount.

The FFA triangle is about prioritization. It is implied by the classic project triangle, but it serves a somewhat different purpose. This is all about trying to figure out which features are the most (or least) important.

Which leads us to this triangle - the bad news triangle. This is the one that gets pulled out when hard decisions have to be made, the one that people most want to deny. (Deny it all you want - this is like the Law of Conservation of Energy at work here...). At a certain point, something has to give - it can be that the schedule might not happen exactly on the target date, or it might be that the budget might be a bit different than originally anticipated or that the features delivered might be a bit different than originally anticipated. Maybe this should be called the Honesty Triangle.

UPDATE: Some people (who didn't bother to leave a comment) feel that the last paragraph is too negative, that it implies that projects always ride off the rails, and that drastic scope needs to be cut, or drastic schedule slippage must occur or that massive budget adjustments must be made. Sorry if you got that impression - that's not what I'm trying to say.

The point is that there MUST be a flexible point on the triangle. The triangle is viewable through various perspectives, depending on the condition of the project, which can be influenced by many, many factors. No one wants to be in the Honesty Triangle, which is the point at which you must make hard decisions. And most of the time, you don't have to be there. But, should you, for whatever reason, find yourself with a project where the schedule is slipping, then its time to look at the honesty triangle. Honesty hurts, remember.

Distributed Art

Once upon a time I had a rock band called Blue Shift. We made music, loudly, in smoke-filled bars (remember those?) in the 90's in Toronto, wishing we had enough money to be able to afford to record more of our stuff. I like to think that we were ahead of our time. As you may have noticed, you've never heard of us.

A little while ago I was suddenly struck with a revelation - a lot of the problems I had back in those days, stuff that held us back, had simply packed up and left in the night. I could now afford recording equipment, due to both my increased income (from, you know, zero) and by the fact that digital recording technology had made professional recordings so much cheaper than they used to be. There were many channels to digital distribution open on the Internet.

I had been sitting around thinking about trying to scrape together yet another band from craigslist when suddenly it struck me - why not just go and get the old band? The fact that we don't live in the same city doesn't matter.

So now I'm looking at four words I never, ever, thought I would see: "New Blue Shift Album". We're all sitting on our own project studios, as it turns out. The project will basically work like this:

• We'll agree what songs to record. We'll establish song structure and tempos.
• I'll record a scratch vocal and guitar track to be a guide.
• Our bass player will be doing drum programming - he'll create bed tracks (drum and bass).
• We will lay down our overdubs (guitar, vocals, keys etc).

Any one of us are allowed to mix at any time. Someone doesn't like my mix? Fine - make your own. We continue to lay down tracks and make mixes as we go along.

Eventually we agree on approving mixes to be the "official version". We get enough tracks like that - and that's the new album.

I am quite inspired by this - this is a way in which the Internet has personally changed my life - something that simply was impossible before is now in reach. I'll probably throw up a side blog to talk about this project as it progresses.

Meditations on Tech Conferences

I went to PyCon over the weekend. It was definitely the largest tech conference I've been to so far. Subcultures fascinate me in general, but this was really interesting. Over 1000 really smart, focussed and innovative technical types crammed together in the same hotel. It was kind of like drinking from a fire hose - I needed to take little timeouts in order to make it through the weekend.

Geek culture is generally polite and trustworthy. I found that I had no problem letting a complete stranger watch my laptop while I went to the bathroom.

Maybe it was just affinity at work. I sense that many people were dealing with similar problems that we were trying to solve in different ways. For example, I found that one tutorial leader was applying himself to a problem space (massive, fail-safe parallel computing), using Python, that seemed very similar to the kinds of things that Erlang was designed to handle.

Perhaps Test Driven Development should be called Guilt Driven Development. Mostly it seems to just make developers guilty about not testing their code enough. Or for those who do test, smug. Bastards.

Oh and finally, I learned a lot about using slides. So many people know this, but I've never seen "less is more" so graphically spelled out for me. I'm never using bullets on a slide again.

Marketing Smarts II

Not to get repetitive, but SeamlessWeb is at it again:



The kicker is that SeamlessWeb doesn't email me much. Once a week, tops. So when these messages arrive, they command my attention. The weather messaging in this context is note-perfect.

Ai Fridays

Today marks the start of Ai Fridays, a new way for us to wind down the workweek. This is a hard-working office, and we're doing our best to acknowledge the need for a) targeted, stress-free time and b) a little bit of fun.

Our new Friday format is threefold.

1. Breakfast in the office. Some of our peers buy everyone lunch daily, but that's more of a perk, and this is more of a thank-you for a job well done. It's also a nice way to start the day. The breakfast station behind my desk has been convivial all morning. Our routine will be bagels, orange juice and the like, although at an unspecified Friday in the future I'll be rolling in an omelette station.
2. Fuss-Free Friday. This applies to how we're supposed to work: no meetings, no calls, no instant messaging, infrequent email. In a client-driven business, we often spend the bulk of the week in conversation, leaving little time to get things done. Fridays are now earmarked for "me" time, where we can all focus on projects with minimal distraction.
3. An early whistle. At 4 or 4:30, we're turning on music and taking the beer out of the fridge. Ai employees typically work past 6, so the last few hours of the week are now a weekend kickoff. We'll get a little work done, but we'll get to enjoy some communal downtime and finish Fridays with a smile.

This isn't particularly innovative; it's just a nice way for us to wind up each week. It's also year-round, unlike the summer or casual Fridays found in many offices. (Indeed, I'm not sure we can get much more casual.) Mostly, it's a way for Ai to show its appreciation to the team, with a little added productivity as well.

I'm posting this on a Friday, so if you'll excuse me, I have to grab a bagel....

Silicon Alley is not back

I'm not terribly consistent with blogging or speaking to the press, but this past week I've done both. I weighed in on an article that Tom Acitelli writes in The New York Observer asking if silicon alley still exists.