Zen and the Art of Programming

People Driven Development

2013-03-24T01:28:45Z

At every stage of the software development process, I like to put people first.

I’m deliberately using the generic word people instead of the more common users. In many cases the decision will indeed affect end-users, because designing a good user experience (UX) is fundamental to the production of quality software; however my point is to try to consider all the people who may be affected by a given decision.

For example, a choice you make might affect fellow developers in your team, testers, the person that will eventually write the manual for your software, co-founders who’ll have to present the project to prospective investors, or even the sales person that demos and sells the product.

Putting people first each step of the way, or taking a People Driven Development approach if you will, is in my experience a great way to ensure the quality of your product.

As developers it’s easy to be so bogged down in implementation details that the focus on what really matter is lost. What matters is people.

An In-depth Review of Fitbit One, Fitbit Aria, and Fitbit Premium

2013-01-09T00:36:18Z

According to Wikipedia, the Quantified Self is “a movement to incorporate technology into data acquisition on aspects of a person’s daily life in terms of inputs, states, and performance (mental and physical)”. I find data to be a catalyst towards improving my behavior and habits, so I decided to quantify an area of my life that I have often neglected in favor of others: my fitness level (or lack thereof).

I looked into several devices that could aid me with my goal, and found that the two most accurate, appealing, and convenient trackers were Fitbit One and Jawbone UP. The latter’s most recent version is slated as “coming in 2013″. Cool as the Jawbone UP is, I opted for the Fitbit One due to its discreet and comfortable nature, as well as the reputation of the company behind it.

If you prefer a bracelet (much like the Jawbone UP) to a clip on tracker, Fitbit recently announced that you can now pre-order an alternative to the One in the form of a bracelet/wristband form called Flex. Same features, different format. It will be available to ship come spring 2013.

While reading reviews about the Fitbit One I noticed that the same generic information was being rehashed all over the place, so I decided to write a very in-depth review myself. In this post I’m going to tell you everything you need to know before deciding if this device is for you or not. The bad, the good, and the ugly.

The products I will be reviewing today are:

Fitbit One: the activity and sleep tracker.
Fitbit Aria: a wi-fi enabled scale that automatically records your weight and body fat % on the Fitbit.com website.
Fitbit Premium: a paid section of the otherwise free Fitbit.com website. (This link brings you to a side-by-side comparison of the free and paid services, if you’d like to see a comparison of the features.)

Fitbit One Review

I ordered my Fitbit One from the official site. They offer free shipping, but I opted for the standard (inexpensive) shipping, as I was very eager to start using my Fitbit, and it reached me here in Canada in four days (in the midst of the holiday mail rush). Check your local Apple or Amazon site if Fitbit won’t ship to your country.

What’s in the package?

When my Fitbit parcel arrived it contained the following items:

The Fitbit Tracker, which is surprisingly tiny. I got the black one with a blue display, but there is also a gorgeous burgundy one with a pink/purple display, which may appeal more to some women.
A clip on holder that enables you to clip your tracker to your pocket, belt, or bra.
A wireless dongle that allows you to synchronize your activity and sleep data with your computer.
A USB charger.
A sleep band that you put on your wrist at night before going to sleep.

Fitbit is a company that understands what a polished hardware product is. Everything, starting with the packaging has an “it just works”, Apple-like simplicity and feel. After unwrapping your Fitbit, you need to charge your tracker (charging is really quick and should take less than an hour) by plugging it into the USB charger and installing a software called Fitbit Connect on your computer. This software, along with when the wireless dongle, will allow you to routinely synchronize your data.

Tracker features

The Fitbit tracker has the ability to collect the following information:

Steps taken
Distance covered
Floors climbed
Calories burned
The exact time all these metrics took place
An arbitrary activity score (out of 1000)
Hours, initial time to fall asleep, and number of times you woke up during the night

The daily stats for all but the last item can be seen directly on the tracker’s led display before you even synchronize them by pressing on its sole button.

You can also enable alarms (including weekly schedules such as Monday to Friday wakeup alarms that don’t go off during the weekend) to gently wake you up through a vibration against on wrist (instead of the dreaded alarm nightstand alarm clock). This is a feature I found to be surprisingly nice. The vibration definitely grabs your attention enough to wake you up, but I never feel annoyed by it. It’s a gentle process, and above all else, it doesn’t wake up my wife, who is an incredibly light sleeper, like an alarm would (I work in EST while living in PST, so I essentially have to be up by 5:30 am during the workweek).

Accuracy and durability

When you setup your tracker, you create a Fitbit.com profile in the process as well. The procedure asks you personal questions such as your height, weight, gender, age, etc. So the calories burned in relation to your recorded activity level can be better estimated. It’s worth noting that the calories displayed on your tracker include both activity calories and the basic metabolic rate (BMR). Essentially this means that even if you’re in bed still and don’t bat an eye, the tracker will slowly increase to account for the energy expenditure required by your body to keep you alive.

Based on my home tests, the tracker is very accurate when it comes to steps, distance covered and floors climbed.

All in all, I’m pretty confident about the numbers this little device spits out. It’s also very robust for its size, and water resistant. This means that it can come in contact with sweat or the occasional drop of water, and you won’t damage it. Just don’t shower with it, submerge it fully in water, or put it through the washing machine. Other than when I’m showering in the morning, it’s with me at all times.

I found the clip on part to be less durable however, given that the bottom part came off after a week, though it is possible I received one that was defective and was just poorly glued on. My Fitbit is still usable without that end cap, it’s just more likely to snag on fabric as you clip it on. Either way, I contacted Fitbit support and they promptly replied, telling me that a complementary replacement clip on was on its way. That’s good customer care and another notable plus for this company.

Synchronizing Fitbit data, and battery life

There are two ways to send the information the device tracks to your profile on Fitbit.com: through your computer (wirelessly via the USB sync dongle), and through your smartphone (via bluetooth). So far I’ve found that the smartphone route (via my iPhone 4s) works every time.

The same cannot be said for the wireless computer sync which will fail sometimes if the tracker is not in clear sight of the USB port where your tiny wireless dongle is plugged in. (Or it could just be that my belly is in the way ).

When it fails, you’re told so and can try again, which is nice because it means you’re not left with a frustrating “I’m not sure if it synced” type of situation. Nevertheless, I find myself more inclined to use the smartphone option, as it will automatically synchronize the data as long as the free iOS Fitbit app is running.

The estimated battery life of the tracker, even with my multiple synchronizations each day, is about a week. I usually get a warning on the Fitbit.com site that I’m running low on battery power every five days or so. When this happens, I remove it, charge it as I sit working at my computer, and then place it back in my pocket where it’s invisible to onlookers. I find its battery capacity to be both adequate and convenient.

Sleep tracking

Sleep is one of the most determining factor for your health, energy, and mood throughout the day. I know for a fact that I, like many other programmers and entrepreneurs, don’t get nearly enough of it. So I was excited about the ability to track my sleep.

I wasn’t disappointed by the sleep tracking capabilities of the Fitbit One; here’s an overview of its workflow:

Remove the tracker from your clip on or pocket, and slip it into the arm/wristband provided.
Place your arm/wristband on your non-dominant arm (in my case my left wrist).
When you are ready to try to fall asleep, hold the button on the tracker for a second or two. A timer will start.
Sleep.
When you wake up in the morning, hold the button on the tracker for a second or two, as this will stop the timer and thus cease recording how long you were asleep for.

The tracker will log the first time you pressed the button as the time you went to sleep. And the second time your press it, as the time you woke up. Meanwhile it will calculate, based on the movements of your arm, how long it actually took you to fall asleep. It will also record how many times and when (in one minute intervals) you woke up during the night.

Once this data is synchronized on the Fitbit website, you’re able to see the sleep data for a given day, as shown below (my mouse pointer was on a particular red spot to display the time):

Empirically, I found that a bathroom visit at night appears as a series of red rectangles next to each other in the middle of some blue rectangles. A simple waking up, rolling over, and falling back to sleep again will show up as just a rectangle or two in my case.

With this data at hand you can start to spot patterns and correlate your lifestyle with your sleep. Since the site allows you to log mood, you can even compare how your sleep pattern affects your overall mood (keeping in mind that correlation doesn’t imply causation and all that jazz).

By observing your sleep data, you may be able to answer questions that you have such as, does drinking coffee during the day affect how many times I awake at night? Whether I get a fairly uninterrupted night of sleep? How long it takes me to go sleep? What about reading at night? Or exercising during the day? And so on.

The obvious shortcoming is that if you’re really tired, you might fall asleep before you press the tracker button that tells the tracker you are ready to fall asleep in the first place. Thankfully you can log sleep manually if you remember the approximate times for when you went to sleep and when you woke up, (the red spikes within that range of time will still be shown as times when you woke up, because your movements are recorded whether you’re in sleep mode or not).

iPhone app

I would rate the iPhone app as a 6/10. It’s bare boned and gets the job done (most of the time), but the cool reports available on the website are nowhere to be found on the app.

I think the basic idea is that you can use the app as a way to track food or synchronize your data on the go, but if you really want to do some serious analysis, you’re better off grabbing a computer.

Truth be told, I also find the iPhone app somewhat slow at times, particularly when I haven’t retrieved data from the site for a while. We are talking a few seconds tops for the wait time though, nothing too annoying, but I have seen more responsive apps for sure.

The food tracking section of the app doesn’t have the ability to scan food labels like many weight loss and food logger apps do, which is certainly a shortcoming.

Sadly there isn’t an iPad app either, which could really shine given the greater real estate that an iPad screen affords. Instead you can use the iPhone app on the iPad and enjoy the eye straining or pixelated (at 2x magnification) experience. It’s a missed opportunity for Fitbit, and one that we can hope they’ll pick up on in the future.

While I’m happy that there is an iPhone app at all, I find it to be one of Fitbit’s weakest links. (An Android App and a mobile site also exist, but I can’t comment much on these as I haven’t personally tried them.)

Conclusion about the Fitbit One

I absolutely love my Fitbit One and I have already recommended it to a couple of family members and friends. So far both my mother-in-law and my friend Marco have purchased one, and they immediately fell in love with theirs.

Of the three products I talked about here, to my mind, the Fitbit One is the most indispensable one. At $99.95, it’s worth every penny, and I think most people would enjoy — and could benefit from having — a greater awareness about their activity level throughout the day.

Fitbit Aria Review

I went all out with my order from Fitbit. I got the tracker, and also decided to buy their wi-fi scale. It’s gorgeous. If Apple designed scales, this is what they would look like.

The setup is done through software on your computer, so there aren’t any buttons on the scale and don’t have to enter your wi-fi password with your feet.

You can configure it to recognize up to eight different registered users. Once you step on the scale, it shows you your weight, body fat %, and your initials. If two or more people using the scale have a similar weight, the scale will show the one person it guessed and allow you to tap it with your foot to select a different user if its guess was mistaken.

After the weigh-in, the scale will automatically use your internet connection to communicate with Fitbit.com and update your profile with your weight and body fat % for the day.

Realistically, if you already have a quality scale (which I did) this is more of a convenience than anything else. You can very easily weigh yourself with your existing scale and then manually log that date into the Fitbit system either on their website or through a smartphone app.

Perhaps surprisingly, I ended up giving the scale away to my lovely mother-in-law who was considering buying one (and who by the way, hardly needs one as she is in great shape), just two weeks after purchasing it.

My mother-in-law likes the scale a lot, however, I did not. I appreciate the beauty and simplicity of the scale. I also really enjoyed the convenience of stepping on and having your weight recorded automatically. What I didn’t like, and the main reason why I gave it away, is the fact that I didn’t find it to be accurate.

Let me explain. When my mother-in-law and my wife used the scale, the accuracy was great. They could step on the scale multiple times and get the same reading virtually every time. It was consistent. When I did the same thing though, being much much heavier, the location of my feet and my balance really affected the pressure value of each of the four springs at the base (that was sitting on hard floor). So depending on how I positioned my feet, as well as my posture and balance, I got different values. And I’m not talking about a fraction of a pound here, I mean up to 3 lbs difference between one reading and another taken just ten seconds later.

As such my trust in the scale was tarnished, and when you are trying to lose weight, it’s unnerving to not be able to trust your scale. Which value do I go off of? The one that said that I lost two pounds or the one that said that I gained a pound? The body fat % estimates, which are already known to be inaccurate on bioelectrical impedance based scales, was also variable (sometimes by as much as 5-10%) between readings.

I don’t care if a scale doesn’t give me a value that is actually my weight to the exact pound. What I do care about is the delta, as it’s harder for me to believe in the numbers I’m seeing and to figure out how the changes I’m making to my diet and lifestyle are having an effect on my weight and body composition.

In my case I have a bad knee due to osteomyelitis (which I contracted at birth), so my weight tend to be distributed unevenly between each of my feet. This might have something to do with my bizarre results. Still, the older scale I have (a Tanita BC-554) tends to give me consistent values between consecutive readings that differ at most by only 0.2 lbs. If I step on the Tanita scale trice, I’ll get two values, if not three, that are identical each time.

Conclusion about the Fitbit Aria

My conclusion is that the Fitbit Aria works best if you are not too heavy and you have good balance. If you’re a heavier weight, I have a hard time recommending this scale to you based on my personal experience. This is a shame because it’s a sleek, beautiful looking scale and a great concept. Down the road, I might buy a new one, or a 2.0 version, once I’ve lowered my weight to below 200 lbs. But for now, I feel less stressed dealing with a scale that I trust.

Fitbit.com Basic and Premium Review

The Fitbit.com website is where you get to analyze your data in all of its glory. It’s also the easiest place to log all sort of measurements and, of course, the food you ate (if you decide to track calories).

The basic site is free, but there is a Premium section with extra reports and tools. A very short free trial version of the Premium service is available (it only lasts for a day for most reports), but let’s first take a look at what the free version has to offer.

When you log in on the site, you’ll be directed to the Dashboard which prompts you with key information about your stats. By default most of the information displayed pertains to the current day and your progress in relationship to your goals. By default you’ll have goals set by the site which can be customized to your liking, as I did, for example, for the number of daily steps (which I reduced from 10,000 to 3,000).

You can select if you want to see the same information on an arbitrary date.

The chart at the bottom of the first section of the dashboard, also reveals interesting information about the amount of activity you did in a given day.

You hear the words sedentary and active tossed around rather lightly sometimes, however here you actually get to quantify them in relationship to your daily life and activity level.

The dashboard also include a section for your Food Plan, something that allows you to specify how many pounds you want to lose and how aggressive you want to be in doing so (there are four intensity levels spanning from easy to hard).

Once it’s set up, the dashboard will tell you how many calories you have left to eat for the day (or how many you ate on a previous day, if you want). This number automatically adjusts based on your activity level, so as to ensure that you don’t go too “underboard” on the amount of food you consume if you are working out really heavily.

The dashboard can also display statistics for the week (which are also sent by email), month, or even year. For example, the weekly view of your Food Plan section in the dashboard will appear in the form of the following type of calories in vs calories out chart.

Make sure you disable “Calorie estimation” in your account settings. Otherwise during days when you log very little activity (e.g., you spent your day sitting at your desk for twelve hours), the site will assume a default average sedentary lifestyle (as it thinks you forgot to wear your tracker for a while), which is actually higher than the level of activity you did. So the daily expenditure will be estimated to be higher than it actually was. This happened to me one day. It was my “laziest” day ever while using Fitbit and yet the calories reported were on par with some of my most active days (yes, that tells you that I’m pretty sedentary).

Aside from all these details and many others (your weight, body fat %, sleep, etc), the dashboard features badges when you achieve milestones, as well as a friendly competition with your friends, based on the most logged steps in the past 7 days. The My Achievements tab will also tells you lifetime stats and personal records (e.g., the day you walked the most steps).

Friends only get to see the details you intend to share. And that can range from as little as your daily steps all the way to making your entire profile available to them (or even to the public at large, if you’re so inclined). The site gives you full control over your privacy settings, as well as sensible defaults (in the past these defaults weren’t too sensible).

I find the gamification aspect of the site to be very compelling. You really want to push yourself to surpass your friends, while collecting badges for both activity and weight loss (assuming you’re trying to lose weight).

Joining groups

The dashboard will also display a list of the groups that you belong to for quick access. These thematic communities (e.g., Canadian users) provide a discussion forum and a leader board of the most physically active participants in the group (as well as your ranking amongst group members). Sadly, there is no search function to find groups (short of using external tools, and the groups aren’t indexed in Google either, to further complicate matters), so you are forced to go through pages of groups starting with !!! in their name in an effort to appear first. Joining groups is frustrating. This is a major oversight by the Fitbit team, and one that should be addressed as soon as possible.

A complete list of communities can be accessed via the Community tab at the top.

Logging data

Aside from the Dashboard and Community tab, the site has a free Log tab. From there you can log food, additional physical activities that aren’t recorded by the tracker (e.g., swimming, cycling, or weight lifting), weight, body measurements, sleep, mood, allergies, journal entries, heart, blood pressure, glucose, and custom metrics of your choice.

Along with logging, these sub-tabs act as reports of sort with charts and useful information.

Realistically the charts get the the job done, but I have seen prettier and smarter ones on other sites before. I’d like for example to see a trend line and have the chart zoomed in on the area where the current values sit.

Logging food

Logging is generally all fine and well, however the Achilles’ Heel in this case is the food logging system. The database is pretty limited compared to other applications and websites of this nature, so you’ll have to manually enter quite a few foods (depending on your eating habits of course). The US database also contains different nutritional information for some products that are also sold in Canada. So sometimes I still need to enter a custom food even though the same product name and brand already appears in the system.

The most annoying part when it comes to food logging is that you need to enter the name of the product carefully and in a certain order or the search function will occasionally fail to show you the most obvious product. I would expect the results to place user entered foods at the top, but they don’t. So if I enter “Dark Chocolate” I won’t even see the delicious 88% Endangered Species Dark Chocolate entry I’ve created earlier on at all. There are favorites, recent, and custom foods lists from which to pick from to speed up the process of logging foods, but it would be nice to have a larger database and a more polished search feature.

Some users opt to use the free (excellent, may I add) site MyFitnessPal (MFP) and synchronize the information they log there with Fitbit.com. In practice, you log foods on MyFitnessPal and the nutritional information is sent over to Fitbit.com. In turn, data regarding your weight, body fat %, etc, gets taken from Fitbit.com and placed into MyFitnessPal as well.

I linked the two sites, but I don’t bother logging food on MyFitnessPal, as I prefer to use a single site (i.e., Fitbit’s). Others might find Fitbit’s logging system too frustrating and thus welcome the MFP option. If you do, keep in mind that MyFitnessPal’s huge database is user contributed and as a result sometimes has incorrect nutritional information. Double check how many people confirmed a given entry before blindly trusting it.

Where Fitbit absolutely nailed it is the way they handle different units. Very often you can just specify the quantity in your preferred unit (e.g., g, oz, tbsp) and the system will do the proper calculation for you. In fact, when you specify custom foods, you can provide alternative units that are not directly convertible. For instance, you can provide the nutritional information for both 50 g and 5 slices of a given ham, and you’ll be able to specify grams, kg, oz, or slices next time you want to log that food.

The Food Log section also provides you with a bare bones report of what your macronutrients are and what you ate at different times throughout the day.

In the sample screenshot above you’ll notice net carbs. That’s something I added in with a basic Google Chrome extension I created, as I’m eating low carb these days and need that information. If you are not on a keto, primal, or low carb diet in general, you probably won’t need my Google Chrome extension. Stock Fitbit.com doesn’t have Net Carbs (which are the total carbohydrates minus indigestible fiber) in the daily total, and by the same token it has Sodium instead of Net C. in the table.

Premium features

The Premium tab has six sub-tabs: Benchmark, Food Report, Activity Report, Sleep Report, Trainer, and Export. These optional features are available for $49.99 a year.

Benchmark compares the number of steps performed against other users in similar demographics (and among all demographics).

Food Report gives you a wealth of information about your weekly macronutrients. (The image below is intentionally shortened with the black band in the middle.)

Activity Report is like Benchmark, only based on calories. This report also includes details about how you’re meeting the goals set by the automated trainer program. The trainer program is activated through the Trainer tab and helps you increase the number of calories you burn on a weekly basis, by increasing this goal gradually each week.

Sleep report does the same comparison with other users, but based on sleep time. The blue area tells you the recommended amount of sleep you should be getting, and the green bar points out the typical amount that other Fitbit users got. It also provides details on whether your sleep levels are improving or getting worse over time.

Finally, the Export tab allows you to be in charge of your own data by exporting it and then analyzing it with the software of your choice.

Conclusion about Fitbit Premium

The reports, export, and trainer are nice features that enrich the Fitbit experience. Are they indispensable? Not by a long shot. But in my case I find them to be worth the less than $5 a month (paid yearly) that Fitbit charges. For the same price of a Starbuck visit, which I’m not doing anymore to begin with, I get an extra boost of motivation.

In my case I find the statistical distribution for my demographic to be very telling of how much I need to improve my fitness level. There is a certain sample bias at play here, given that you’re amongst a group of people trying to improve themselves and not the general public. Still, I find that being so much on the left hand side of the distribution has been a strong motivational factor to improve and slowly reach a point where I’ll do better than the typical user on all the performance variables that matter.

Likewise the trainer has been a great feature for me, as it promotes behavior altering thoughts. For example, now I’ll park my car further than I used to, so as to get in those extra steps. I will happily shop with my wife for hours on end, and will take the longest route possible when walking somewhere, so as to do better in these stats.

In fact, the following graph from last week tells an interesting story.

As you can see I wasn’t meeting my goal for the week, so I really gave it my all on Saturday in order to compensate. If the trainer wasn’t in place, I wouldn’t have had a reason to try to go the literal extra mile.

General conclusion

In conclusion of what is now a huge review, I’d like to point out that as cool as data is, awareness without action is pretty pointless. I’m finding the whole Fitbit ecosystem to be a catalyst for my weight loss and overall fitness improvement.

Over the past month I’ve lost over ten pounds, gotten my blood pressure and heart rate to a desirable range down from some rather dangerous levels, and also been able to drop my blood glucose drastically to, practically, within normal levels without taking any medicine. Things are improving and doing so quickly. I attribute 80% of the merit to my diet, and only 20% to the increased activity level. Thankfully, Fitbit has been helpful with both for sure.

The most important tool is always going to be my willpower and a desire to pursue a healthier lifestyle, but I find Fitbit to be a rail upon which my hand can rest as I embark on this huge uphill climb. If you are in the same situation or simply want to improve your health, I highly recommend giving Fitbit a try.

Note: Links within this review include my referral id. I do not let this affect my judgment of the items and services I review. Whenever I felt there was a shortcoming or issue, I had no problem pointing it out for the sake of my readers.

The Good Enough

2012-11-28T17:08:59Z

A couple of weeks ago Dustin Curtis published an article called The Best. Within it he argues the merit and freedom that one can obtain by spending the time and money required to purchase the best products. The idea is that if you can trust things you buy, these items become invisible to you in way because you don’t have to worry about them failing to do their job. This article became popular on Hacker News and many commenters seemed to agree with his view point.

Then along came The Worst, a post which pretty much dismisses Dustin’s argument as consumerism, and proposes to take the opposite strategy to achieve the same goal of not worrying about stuff. Buy the cheapest items that you don’t give a crap about, so no matter what happens to them, you won’t care.

My personal philosophy has always been one of “good enough”. In short, I don’t have any hangups or philosophical issues with purchasing cheap or expensive things, and unlike the writer behind The Worst, I don’t have too many problems with consumerism either. I take the following, arguably pragmatic, approach to this area of my life.

My decision process typically involves the following parameters:

Price of the worst, middle ground, and best options available
Estimated difference in user experience between these three “levels”
How often will I use the product
How serious the consequences are if the product fails to perform its duties
Useful life expectancy of the object

Using these considerations leads me to avoid purchasing ‘the worst’ in most cases. The difference in price between the worst and the “good enough” is usually not huge. The quality, reliability, and features, however, usually are. The absolute cheapest products tend to have poor performance and reliability, so what I end up buying most of the time is something that is not quite the best (with its ridiculous price tag), but rather something that’s sufficiently good enough.

I buy $100 headphones, not $1000 like an audiophile might. I’ll have a nice $50 per person meal at a restaurant, but I won’t unnecessarily indulge in a $300 binge at the priciest joint in town. I will buy a $500 compact camera, but not a $2000+ Leica X1.¹ I bought a $80 tower fan this past summer, but not a blade-less Dyson. I have my eyes on the upcoming ~$2000 Canon EOS 6D, but I won’t splurge for a $7000+ 1Ds Mark III or even a $3,300 5D Mark III. I don’t need Dustin’s $50 per set cutlery², I’m happy with a $5-10 setting and a KitchenAid steak knife for an additional $5. You get the gist.

At times I will buy the so called ‘best’, but it’s only for selected items I truly care about (e.g., my smartphone, my laptop). These items are fundamental to my day-to-day life and what I do for a living.

If I don’t care very much, I won’t spend time researching what’s good enough either. I’ll use price as an indicator of quality and shoot for the middle or just below it. The method is not foolproof, but it generally works for items that don’t really matter to me. Why spend scads of time obsessing over what’s the best item if it’s something that I’ll rarely use, or whose failings won’t affect me too much, or that I’m likely to replace in a year or so anyhow.

Psychological research tells us that we’d be happier to buy products without the due research and countless hours spent reading reviews and specs online. This is true, because the more you read, the more you realize even your “good enough” product has flaws and shortcomings. It’s draining at times, but after the purchase is made, and my expectations have been lowered a bit perhaps during the research phase, I often find myself very happy with the quality of the products I chose.

One case in point is that of our hand vacuum. We first bought a ($65) hand vacuum that died in about a year and which could barely keep a charge for a minute. We then replaced it with another one which had incredibly poor suction (to the point where it truly was unusable). The third vacuum lost suction power after 30 seconds and would only keep a charge for two or three minutes (a combo of both of the previous model’s problems). We were frustrated by how this seemingly not very significant household kept failing to preform properly, so I took the time to do more research online and discuss the matter in person with an expert vacuum repairman. We now have a very powerful Eureka 71B, which interestingly wasn’t more expensive than the ones we had already bought in the past. It may not be the absolute best in the wold wide world, but it’s good enough.

Likewise, when I was in the market for a car, I did extensive research and I ended up buying a Jaguar. All the time spent researching a safe, comfortable car paid off and I couldn’t be happier with my vehicle and its performance. It might not be a Lamborghini or a Bentley, but it is undoubtedly good enough. The research and investment was worth it for such a crucial purchase. The consequences of a bad choice here could lead to serious headaches and, in the case of an accident, it could make the difference.

I get most of Dustin’s benefits and comfort without spending as much time obsessing over relatively unimportant things like cutlery. And therein lies my approach, flawed or not, to acquire useful things that make my life simpler, more fun, or simply aid me in some capacity without necessarily splashing out great sums of money on everything I purchase.

[1] Disclaimer: This and other links within this post have my Amazon Associates ID. I get a small cut, at no expense to you, if you end up buying from Amazon.

[2] Nevertheless I have added them to my wish list as I’m curious now to try them.

A Review of the TV Show Start-Ups: Silicon Valley

2012-11-14T17:47:05Z

Several years ago I remarked to my wife about how mainstream media was largely ignoring the world of startups. The amount of coverage in books, news, television and movies was sparse for such a revolutionary industry, let alone one with a track record of producing billion dollar companies.

In some ways it still is, but things have slowly improved, as I conjectured at the time. Startups have garnered much wider mainstream coverage, the news has an unhealthy fascination with Twitter, and you can’t walk in a reputable bookstore without finding several titles about startups.

At the time, I also mentioned that I’d be very interested in watching TV shows and documentaries about startups and other high-tech companies. If you’re running your own startup or are into the startup world in some capacity, such productions would be definitely interesting, useful, and entertaining.

Thus I was very excited when I discovered that a new show about Silicon Valley had kicked off on Bravo. The show is called Start-Ups: Silicon Valley and has none other than Randi Zuckerberg (Mark’s sister) as an executive producer. (She joined the team at a later stage, so she was not involved with selecting the cast.)

I must prefix my review by mentioning that I am not a TV snob. I watch a wide range of shows, including top-notch programs like Boardwalk Empire, Mad Men, Dexter, Breaking Bad, Top Gear, but at the same time, I don’t mind some of the sillier, “just for fun” ones either (e.g., South Park, Family Guy, The Big Bang Theory, and How I Met Your Mother), as well as some reality shows, too (such as Pawn Stars, American Restoration, Storage Wars, American/Canadian Pickers, and Tanked). I don’t watch shows like Jersey Shore, Big Brother, Honey Boo Boo, and so on, but I won’t shy away from the occasional episode of Maury if nothing else is on.

That is to say that I can appreciate a multitude of styles and am okay with reality TV being somewhat scripted for comedic or dramatic effect.

I watched the first episode of Start-Ups: Silicon Valley recently and my initial reaction was, “Are you freaking kidding me?”. Nobody expects reality shows to accurately represent reality, but I feel that in this case, they grossly missed what it means to run a startup. As well as bypassing most of the culture that makes the tech world unique.

It’s hard to generalize, as there are all sorts of people running startups, but you can rest assured that the majority of them are up late writing code, not frolicking at glamorous toga parties as depicted in the first episode. The misrepresentation doesn’t end there however. Not by a long shot.

The cast members are not your typical group of startup folks. For starters, not many of them are actual programmers. They most prominent ones are depicted as social butterflies trying to use their connections to succeed. Who you know does matter, but it’s definitely not the focus in the disruptive environment of startups. However the show portrays this as pretty much being the only goal and requirement one needs to succeed in Silicon Valley.

Critical components of running a startup, such as customer discovery and acquisition, implementing a great product, and ramen profitability are practically ignored throughout the show. Instead we’re shown a world where startup founders are all extremely good looking and often in various stages of undress. They party, have cat fights over an email (see, they’re so nerdy), get spray tanned, spend hours getting ready, mostly sleep in glamorous places, drive exotic cars, discuss dating (Silicon Valley is apparently 1/3 gay, 1/3 geek, and only 1/3 dateable), and try to pitch to investors in atrocious ways (OK, this last point is not unheard of in real life either).

(Incidentally, this is also what Start-Ups: Silicon Valley’s producers think nerdy guys look like.)

The startup people depicted in Start-Ups: Silicon Valley might look like the cast of 90210, but they are not professional actors. And it shows. So most of the scenes come across as contrived, disingenuous, and more scripted than Chumlee’s shenanigans on Pawn Stars. At some point Hermione (one of the good looking kids on the show) heads over to Dave McClure’s office with her brother Ben (whose relationship is, to my mind, depicted with somewhat incestuous undertones throughout the show).

She is wearing tight Tetris print leggings (see again, so nerdy!) and ends up sleeping under Dave’s boardroom table before the pitch (because she’s exhausted from being up partying the night before). Dave of course randomly wanders into the office to find her sleeping there.

So you are either a completely unprofessional moron who got away with too many things in life because of your decent looks, or this is bizarrely scripted to show how relaxed, laid back, and easy this whole startup game supposedly is. And when Hermione and Ben inevitably get turned down (they couldn’t articulate what their company does in 30 seconds and reveal that they are involved in another 42 startups), one cannot help but sense that the desk and/or similar future incidents will no doubt be used for some extra drama between the two siblings.

There is being creative, hustling,and thinking outside of the box, and then there is being inappropriate and childish. Did I mention that Hermione is not the most surreal character? Wait until you meet her ex-friend turned nemesis, Sarah, a so called social media expert/blogger who lives in a hotel room, and feeds on artificial drama and tweets for breakfast. O.M.G., in the next episode she live tweets a video during a date with a model. How will he react? Will he be OK with it? Don’t miss that episode. How else are you gonna know if it’s socially acceptable to do so the next time you date a model after a hard day of work at your startup?

Even when computer screens are shown, they’re often a caricature of what an actual programmer’s screen might look like. At some point you can see a terminal scrolling (almost reminiscent of The Matrix), yet upon closer inspection, one sees that it’s just a listing of directory and files with a dir command in Windows to make the setting look more impressive than it really is.

Overall this show is an utter travesty and a complete misrepresentation of what the startup world is really like. It is to the startup world what porn is to real sex. It’s Hollywood’s misunderstanding of what the nerdy neighbors in Silicon Valley are doing.

The sacrifices, risks, low profile, and 70+ hours workweeks of most actual startup founders are mocked and trivialized by a show that portrays a carefree world where you can easily make it if you’re connected enough and can pull the right strings. Where partying and looks come first, and where they are virtually never shown, you know, actually doing any hard work.

Look, I get it, coding is boring to most viewers, but surely there must be a middle ground between the fantasyland portrayed in this show and a more documentary-like approach that captures some of the real struggles behind running a startup and what it takes to succeed in Silicon Valley. There is plenty of real drama when you start running out of savings, VCs are turning you down, and your churn rate is going up fast. So much so, that you don’t need petty high-school fake drama to make a show about startups action packed or interesting.

Come to think of it, imagine a show that actually explained to the general public, Mythbusters style, what churn rate, SaaS, and other startup terms mean as they go about following and documenting the journey of a few founders.

Aside from the missed opportunity of producing something entertaining and useful, this show risks further propagating the myth that launching and succeeding at a startup is easy. And that is something we as a community certainly don’t need.

Going into it, I had hoped that Start-Ups: Silicon Valley would be akin to, at least, American Restoration meets the startup world. In reality however, it’s a perverse sprinkling of scripted scenes and a farce of what the life behind most startups truly is. So that when all is said and done, what we’re left with is a Silicon Valley version of The Hills.

Why I Won’t Be Publishing New ‘Great Ruby Shootouts’

2012-10-16T15:15:01Z

Someone on Twitter politely asked me when I was going to run a new Ruby shootout. I replied that I wasn’t planning to. This got me thinking though that he might not be the only reader wondering about this point, so I thought I’d share a few of the reasons behind my decision not to carry on with this series.

My last shootout (over two years ago now) and its predecessors were highly popular posts. In fact, benchmarking in general tends to attract a lot of attention. It wouldn’t really cost me much to prepare a shootout every, say, six to twelve months. The reason I have not done so again is that I’ve come to view these posts as not being of much value to the community.

You can use as many disclaimers as you want in your report, but many readers will still just take a glance at a couple of graphs and draw broad (and not particularly accurate) conclusions.

An implementation could specifically target these benchmarks, optimizing for them, without a real, general performance improvement. Or more often, be so incomplete that doing well in micro-benchmarks becomes almost trivial. (That said you might still see other benchmarks from time-to-time on this blog, when appropriate.)

Even when it comes to complete implementations, a good performance on synthetic micro-benchmarks only gives us a vague idea of how well an application will be run within the given implementation. Why not test your own code with the major implementations you are interested in using instead?

This actually bring us to another point. These days you can easily install multiple versions of Ruby via tools like RVM or rbenv. So if you do intend to test your application with Ruby, JRuby, and Rubinius, for example, you could do so easily and painlessly. And that’s really the best way to choose an implementation (at least as far as performance and ensuring compatibility are concerned).

I understand that there is some value in showing a ballpark comparison, however arbitrary, but I also feel that it tends to separate, rather than unite, the community. All other implementations, but the winning one, end up worse off.

The various Ruby implementers are welcome to continue running these tests if they’re useful to them (keep in mind that the code is mostly a mix and match of very old tests and therefore quite messy). And of course, other bloggers are welcome to run and publish their own comparisons if they feel that doing so is worthwhile. But for all the aforementioned reasons, I don’t plan to continue holding shootouts here.

Join the Upcoming IBM Big Data Developer Day in Boston

2012-09-28T00:22:07Z

Following the success of the first Big Data Developer Day back in June, IBM has organized another Big Data Developer Day. This one will take place on October 10, 2012 in Boston. If you’re in the area, I highly recommend that you register and attend the free full-day event.

You’ll learn more about Hadoop, Social Media analytics, real-time text analytics, and other cool technologies that IBM has been working on for the past few years. Free breakfast and lunch will be provided as well, so there’s all the more reason to attend. If you’re interested, be sure to register here.

If you can’t attend this one because it’s too far away from where you live, rest assured that more events of this nature will be organized in different cities in the future.

Two Big Data Events in Silicon Valley Next Week

2012-06-07T15:49:10Z

Yahoo!, Hortonworks, and IBM are three companies that are heavily invested in Big Data. If you’re in Silicon Valley next week, you should check out a couple of events that they’re a part of.

2012 Hadoop Summit on June 13th-14th
Big Data Developer Day on June 15h

The first is organized by Yahoo! and Hortonworks; the latter by IBM.

The Hadoop Summit is the bigger event, and even if it’s not free, at $599-$700 it’s a must-go if you are interested in Big Data technologies and happen to be lucky enough to live in the Valley or can make it there. For the Hadoop Summit’s agenda check out their busy schedule.

The IBM event is smaller and shorter, but still well worth your time — plus it has the added bonus of being free (complete with complimentary breakfast and lunch).

IBM believes in, and placing its bets on, Big Data, so this is a good opportunity for them to provide you with Hadoop information and a demonstration of some of our coolest technologies in the Big Data analytics space.

No worries, you won’t be sold a time-share condo in the mountains or anything at all really.

The following is the agenda for the Big Data Developer event.

8:00 AM – 9:00 AM	Breakfast and networking
9:00 AM – 9:15 AM	Opening by Anjul Bhambhri, Vice President IBM Big Data Development
9:15 AM – 12:00 PM	Interactive discussions and live demonstrations on Big Data, social media, and log analytics
12:00 PM – 1:00 PM	Lunch and networking
1:00 PM – 3:00 PM	Interactive discussions and live demonstrations on Hadoop, stream analytics, and federated discovery of Big Data
3:00 PM – 6:00 PM	Hands-On Lab with Hadoop Scripting and text analytics

Seats are limited and going fast, so please register now if you plan on attending.

New jobs available at IBM in Toronto

2012-06-05T15:00:30Z

My team is making three new programming roles available (specifically in Markham, Ontario, Canada):

Two 16-month internships for students
A 6-month contract (renewable)

Both the internship and the contract role have a strong web development foundation (PHP and/or Ruby on Rails preferred) as their main requirement.

The contract role also requires that you are legally allowed to work in Canada. For the internships we are more flexible and will accept candidates worldwide, but we still have a strong preference towards local candidates due to the cost and bureaucracy involved with relocating students internationally.

The successful candidates will be involved in various interesting IBM projects that utilize the latest technologies, including Cloud computing, Big Data, Open Source development, and databases.

You can apply through Big Data University if you’re interested.

New Release of the Ruby Driver and Rails Adapter for DB2

2012-05-09T07:52:58Z

The API team at IBM recently released a new update for the ibm_db gem, which includes both the Ruby driver and the Rails adapter for DB2 (and Informix). This coincides with the release of the latest major upgrade to DB2 LUW (Linux/Unix/Windows), which is now up to version 10.1.

ibm_db 2.5.10 adds support for Rails 3.2 and fixes a few minor bugs that have been reported by the community. New features such as Time Travel Queries are supported seamlessly.

The biggest news is the simplified installation process. You now only need to set the DB2′s instance user home, and the gem installation process will take care of determining both which architecture (e.g., 32 vs 64 bit) and libraries to use.

All it really takes now to install ibm_db on a Linux/Unix/Mac box is something along the lines of:

$ sudo -s
$ export IBM_DB_HOME=/home/db2inst1/sqllib
$ gem install ibm_db

Windows users don’t even have to worry about setting up the home directory. It will work both on Windows 32 and 64 bit systems, but unlike with *nix systems, Ruby and ibm_db will run in 32 bit mode only.

To try out DB2 with Ruby/Rails, I recommend you download and install DB2 Express-C for free.

RubyMotion brings serious Ruby development to the iOS world

2012-05-03T17:04:47Z

For the past several months I’ve been involved with, but unable to publicly discuss, an exciting project called RubyMotion. My friend Laurent Sansonetti, creator of MacRuby, has now officially launched a serious alternative to Objective-C for iPhone and iPad development.

RubyMotion is not a bridge; it actually generates native applications that are as fast as the ones written in Objective-C . You can check out Ars Technica’s exclusive article for more details on the project and watch the Pragmatic Studio’s 50 minute free screencast about it to get a better idea of how RubyMotion works in practice.

For a limited time you can buy RubyMotion at an early bird discount rate of 25% off. It’s such a productive and affordable solution, that I’m sure it will quickly become popular among Ruby developers who are interested in iOS development and testing, as well as among existing iOS developers.

Many of the questions you may have can likely be found in the FAQ section, but feel free to ask any additional questions here. Enjoy it.