graysky

Favorite Albums of 2012

2012-12-31T00:00:00-05:00

Despite my lack of writing in 2012, traditions must be respected. Outside of the music itself it was another year where I effectively only listened (and paid) to music from Spotify (some Rdio, etc), and bought a Sonos Play 3 in part to more easily stream Spotify to home speakers. Below are a few of the albums I enjoyed most in 2012, and a Spotify playlist of selected songs from them. This year there was a lot of albums I liked but few I deeply loved.

10. The Shins, Port of Morrow - Their latest felt like a welcome return to the early Shins. "The Rifle's Spiral" and "September" are both solid, but "Simple Song" is the obvious standout. Fantastic song.

9. Sleigh Bells, Reign of Terror - Only a couple tracks, like "Comeback Kid", reach what they nailed on "Treats" but even slowed down their dynamics are great.

8. Metric, Synthetica - They have developed an interesting synth/new-wave-ish sound that keeps it interesting. Not as rocking as "Fantasties" although found its way into regular rotation.

7. Cat Power, Sun - My favorite Cat Power album to date. She's had moments of brilliance in the past and I think this album flows from start to finish.

6. Japandroids, Celebration Rock - It took me a while to get into this album before it clicked. Their energy is infectious. They remind me of what I once loved about Piebald and how they crashed forward.

5. Kanye West presents Good Music, Cruel Summer - I think I have to admit I'm a fan of whatever Kanye puts out. The album has a mix of styles and some of the best lyrics ("I believe there's a god above me / I'm just the god of everything else"). Even features a Ma$e appearance! Oddly one of the duds is the Kid Cudi track which feels recycled.

4. Frank Ocean, channel Orange - Plenty has been written, understandably, about this captivating album. It feels almost sneaky how good it is since it comes in such a smooth package.

3. Electric Guest, Mondo - In reading a few year end round-ups I thought this album deserved better that it received. The Danger Mouse-produced album has his signature style of a mix of throwback and modern production. Bottom line is that it kept my foot tapping and stayed in my head.

2. Cloud Nothings, Attack on Memory - Impressive and solid album throughout, and "Fall In" was my jam for weeks. I would love to see this album performed live, everything has such urgency.

1. Kendrick Lamar, good kid, m.A.A.d city - If Dre likes him, how can I not? Songs like "Backseat Freestyle" and "m.A.A.d city" are amazingly catchy, and the album is surprising in all the best ways when so much of hip hop isn't.

The Solidity of Software

2012-07-02T00:00:00-04:00

This weekend included software failing because of bugs handling the leap second correctly (thanks tidal friction!) and disrupting many popular sites with it. A single change that feels small was able to shatter what had previously felt so stable.

That property has always fascinated me about software. In one momement software can feel like it is made of steel, silently cranking away. You have the feeling it would work for the next 1,000 years if you went away. (Of course, it wouldn't; software decays.) And in the next moment it is on pieces on the floor. Or you find a bug that makes you question "how did this ever work properly?"

This mixture of strength and fragility don't seem to have a simple analogy in the physically constructed world. Buildings don't abruptly crumble, bridges rarely just fall down. It leads me to think that the sense of solidity is at least partially an illusion. That software is less resilient to change and new conditions than we hope, and is in a world in flux.

Favorite Albums of 2011

2011-12-31T00:00:00-05:00

In keeping with tradition, here is a quick summary of music I've loved from the past year. In past years I'd used iTunes playcounts as a barometer for my enjoyment but this year I mostly switched to Spotify so have less data to pull from. I published a Spotify playlist of some of the best songs from the albums included.

10. Real Estate, Days - Definitely sleepy but pretty and at its best feels like it is building toward something.

9. Panda Bear, Tomboy - I really like the sound he has developed - this kind of hazy, surf-y atmospheric is a great soundtrack for a certain mood.

8. Holy Ghost!, Holy Ghost! - Fun, synth-fuelded dance-y 80s pop. The best songs are front-loaded on this album.

7. Foo Fighters, Wasting Light - I miss good rock music. This Foo Fighters album feels like a return to their more straight-ahead rock with less interruptions by ballads.

6. Iron & Wine, Kiss Each Other Clean - Continuing his more textured sounds this album from early in the year planted songs like "Rabbit Will Run" in my head.

5. Girls, Father, Son, Holy Ghost - Great album all the way through although the faster, upbeat opening 3 tracks are what I find myself replaying.

4. Foster the People, Torches - Goddamn they wrote a catchy set of songs, and I am not immune.

3. M83, Hurry Up We're Dreaming - Sweeping album made me like it despite how close it comes to sounding like songs from The Breakfast Club.

2. Kanye West & Jay Z, Watch the Throne - It took me a while to get into this album I think because I had such high expectations for it. "Otis" got a lot of attention but it was songs like "Who Gon Stop Me" and "Gotta Have It" that won me over.

1. Bon Iver, Bon Iver - Impressive and haunting album that is strong and cohesive even with the variation he employs.

Customer Acquisition in Practice

2011-11-18T00:00:00-05:00

@rseanlindsay
Sean Lindsay There are companies that will fail and look back and wish they'd attended tonight's @FounderMentors customer acquisition event #fmcustomers
Nov 18 via Twitter for iPhoneFavorite RetweetReply

Last night Founder Mentors' hosted an excellent event called "Customer Acquisition in Practice" at Dogpatch Labs in Cambridge. Understanding effective customer acquisition is something I'm particularly interested in. Often when I talk with new founders they are much more focused on product & technology (like I have been) than customer acquisition strategy. When asked how users/customers will find them they mumble something about SEO, social media or "virality" that clearly hasn't been deeply considered.

Sean Lindsay, who started Founder Mentors, interviewed a panel with Brian Balfour of Boundless Learning, Dan Marques of Gemvara, Jeff Steeves of Wayfair and Parker Swift of Vistaprint. Sean Laurence took much more detailed notes of the event, but this is what struck me:

Brian smartly advised start-ups to focus on one, or maybe two channels in the early days. To do inbound marketing or experimenting with Adwords takes time, often more time than originally estimated. Spreading yourself too thin means you are less likely to do it well.
Across the board few were using a lot of tools for measuring channels. The most common cited were Google Analytics and Excel.
While no one was endorsing black-hat techniques for acquiring users, no one was exactly condemning using, shall we say, aggressive techniques. Those early in the SEO/search pushed the rules to get top rankings, or those early to social platforms have tried to ride that wave. General take seemed to be that startups shouldn't be unethical but certainly look for new advantages. And that tracking the world of affiliate marketing can be educational even if filled with shady tactics.
Sean asked about whether making one's product or marketing "viral" was more of a myth or something than could be aimed for. Brian and Jeff both picked up the same thread that the first step is having a strong product that someone would want to share organically. The set of share levers and game mechanics are secondary to that, and should be thought of as accelerates.
It seemed as though at least some of the companies don't include headcount costs into COCA. Brian commented that there is a move toward including all costs (which is what I've seen) to more accurately reflect the total costs. That makes more sense to me when trying to compare channels like SEM to PR, inbound but not something I've had to do.
If there was a disappointment to the evening it was when Sean asked about further reading it was a good but short, well-known list. It feels a bit that those who really live in this world are reluctant to broadcast their tips and tricks. Brian said he had a list of video links mainly to AppSumo. I hope he provides that list to Sean!

One final note is that this seems like a skill-set where Boston has some very smart thinkers, although somewhat hidden from view. To be blunt, the visual design of sites like Vistaprint, Wayfair, TripAdvisor and HubSpot often belie how sophisticated the thinking is of those who run those businesses. Or outfits like Shareaholic or Nanigans that may not be known outside of certain circles that are super-sharp in user acquisition methods. Hope this is just one of many down-in-the-trenches events on this topic!

Startup Ping Pong

2011-11-08T00:00:00-05:00

Photo by notic

If you're like me and enjoy both startups and ping pong then you'll love Startup Ping Pong! The objective is to bring together players of any level who the enjoy the epic game of ping pong. And if you need convincing that ping pong is indeed epic, check out the Penny Arcade series, "Paint the Line".

Currently I'm pulling together the ones in Boston and Ajay of Groupme is running one in NYC. Sign up if you'd notified when the next event is in your city, and follow @StartupPingPong. Of course, we're always looking for sponsors to buy pizza & beer if you'd like to be in the good graces of some startup folks.

After a brief hiatus I have pulled together the details for the 2^nd event in Boston. It will be Tuesday, Nov 15^th at HubSpot - please RSVP if you'd like to attend. The first event was a good time (photographic proof) since there is no better way to take a break from cranking away than solely focusing on hitting a little ping pong ball with a paddle.

Caution for Startup Kids

2011-10-17T00:00:00-04:00

Everyone knows startups are hot. Fueled by a strong seed funding climate and several well-publicized successes there is seemingly a rush to start your own company. I'm a big proponent of young folks (i.e. early 20s) joining an early-stage startups for a slew of reasons. Joining an existing startup is a great way to learn whole business works, have much more responsibility with influence on how a product evolves and often at little financial risk. But when I talk to smart, hungry young people who are gung-ho on founding a startup I want to offer a few words of caution as they contemplate their options.

This is by no means trying to say not to start a company but to have clear eyes about the risks and tradeoffs. For those lucky few who create a successful startup or are on the early team of a huge thunder lizard, congratulations! You're living the dream, your ticket has been punched and you're among the fortunate 1%.

Learning Lessons the Hard(est) Way - My concern for intrepid startuppers is that they will pound their head against the wall learning what doesn't work but not necessarily what does. And while failure can be an excellent teacher, it isn't always. Failing to successfully sell a product doesn't automatically teach how to become a great salesman. If one's goal is to maximize learning then it might have been better served by working with a master of the discipline. At a minimum, make sure you have mentors and advisors that you deeply engage to avoid the first-time pitfalls.
Beware the Burnout - If you're in the "Uninformed Optimism" phase of entrepreneurism it's hard to imagine why others would ever make the choice to work at a boring, staid companies when they could be their own boss on a mission to change the world. The halls of bigger companies have plenty of people who worked at one or two startups where they got burned and they got off the roller-coaster. The true serial entrepreneur is a rare-breed for a reason. Burnout is real even for the caffeine-fueled twenty something. Time is on your side but you'll likely only get to make a few bets, so make them count.
Ending with Capabilities Better than Resume - This applies primarily to "business" people at startups who are founding a company in lieu of a more traditional MBA. When looking for the next gig all the hard work you've done won't necessarily be valued in the way you might hope. What a later-stage company needs is someone who is a kick-ass product person, or killer marketer, and so they discount the less relevant experiences you bring. Protect yourself by developing at least one core strength where you have demonstrated excellence.
Digging a Financial Hole - The most obvious concern before founding a company is how you'll pay your bills. The risk is that many startups take years to break-even and fund-raising is much harder than the headlines on TechCrunch would have you believe. In the mean-time racking up credit card debt and missing out on income can shackle you down the line. Of course keeping your expenses low is crucial, but also know your financial limit so that you can pursue future goals (another company, travel, buying a house, etc).

(Mostly) Good Times for Software Developers

2011-09-06T00:00:00-04:00

Yesterday NPR's All Things Considered aired a segment titled "For Software Developers, A Bounty Of Opportunity" that included an interview with me (and my daughter cooing in the background). The point of the story is that in even an overall weak job market there is strong demand for software developers. Since much of my interview with Curt Nickisch didn't make it into the final piece I wanted to expand on my experience and thoughts on the market for software developers.

The Right Stuff

While the overall picture for software developers is definitely bright, the reality is that it is a subset of technologists who reap the benefits. Currently every company I know is having difficulty hiring, but they are only looking to hire people with the right skills^[1]. And I know some tech people who have faced challenges finding a job perhaps partly because despite a technical background they are viewed as having the wrong skill-set for the role. So despite the intense demand there can be a non-clearing market. I've certainly seen this first-hand when hiring engineers that many of the resumes are far from a strong fit and one wonders about their attractiveness elsewhere.

Skill Security

When I was at IBM I noticed that their decisions for when to hire and when to cut staff was driven by its own internal logic more than external results. Even when the company overall posted strong quarterly results they may, understandably, choose to de-invest in some business units or product lines. The result was that my co-workers there never appeared to feel any more secure by working for a large, outwardly stable company. And certainly those working for large companies that are actively shedding jobs like Cisco or Nokia may not feel confident about the future there.

My one true fear working in software is letting my skill-set become deeply disjointed from what the market values. The coworkers I've been concerned for are the ones who have either become complacent in acquiring new skills or were "team players" by working on an older technology that other companies don't have a need for. That is potentially dangerous. To remain able to take advantage of the premium put on skilled developers requires keeping your skills current and finding opportunities to grow.

These two observations have convinced me — someone who is not by nature a risk-seeker — that working for startups is actually a better and more secure strategy. Startups provide greater transparency about their health and offer more chances to develop the right skills.

The Broader Picture

There is much debate about whether there is currently a tech/startup bubble but Marc Andreessen argues persuasively in "Why Software Is Eating The World" that the importance and reach of software in our lives is only increasing. He also touches on the critical challenge that currently there is only a small segment of the U.S. workforce prepared to join those companies in this "software revolution". At the moment I'm an accidental beneficiary of this shortage but this is a long-term problem that needs to be addressed by our education system. For more Americans and those around the world to participate in this trend they need a solid grounding in math & science and the chance for a quality university education. Too few have them today causing innovative companies to be constrained and Americans to miss out on a rare bright opportunity.

[1] I believe that the market for college hires function differently since they are usually judged on learning potential more than by their current skills, and hiring trends are more shaped by the firm's growth plans.

Hubspot Acquires oneforty

2011-08-18T00:00:00-04:00

I've been with oneforty for 2+ years from early days at TechStars and learned an enormous amount from great teammates, investors & advisors as well as from being in the startup trenches. It has been a time of successes, challenges and every other part of the startup roller coaster. I'm deeply grateful for the opportunity and support that I've been given. Today, I'm happy to share that oneforty has been acquired by HubSpot!

HubSpot is transforming how marketing is done and plans to become the "Salesforce for Small-Business". As I've written before I love that they are swinging for the fences. From the beginning oneforty has aimed to simplify the challenges of social media around finding the best tools, having the right knowledge and, most recently, collaborating effectively. We think the problems we've been working on are well-aligned with HubSpot's bold vision.

To achieve this mission they have been assembling a great team. I'm excited to have the chance to work with folks like Dharmesh, Brian, Yoav and David (among many others) whom I plan to learn from. Big challenges still lay ahead, sounds like fun.

Chess Lessons for Startups

2011-08-01T00:00:00-04:00

Photo by aerokev

When I was young I played a lot of chess with my father, who is a competitive player, and in several tournaments (gee, big surprise I ended up as a technology nerd). While I was never going to be Bobby Fischer or Magnus Carlsen I learned lot from playing beyond just the game itself. I've found that several lessons from chess apply to running a startup.

Focus on the Middle Game

In chess there is more written about openings (the Queen's Gambit, the Réti, Dutch defense, etc.) and endgames (mating strategies with various material) than middle games. Similarly in the startup game more time is spent writing about openings (product launch, raising money, first hires) and endgames (acquisitions & exits, scaling big) than the middle game. It is an understandable pattern because the permutations for a chess game are enormous after a few moves and for startups the middle period tends to be more particular to their situation and what has come before.

The challenge is that most games are won or lost in the middle. While it is certainly possible to hang yourself during the opening or blunder in the endgame the real competition is in the slow grind for positional and material advantage. For startups this middle game can take many forms, such as:

Mixed Results Post-Launch - It's after the excitement of launch has dropped off and sign-ups and engagement are only so-so. Should you press ahead because the product is still immature and users say they need more? Or are you mis-reading the customer's level of interest in a solution to this problem?
Follow-on Financing - The seed money is running low and hitting milestones has been harder than anticipated. You've learned a lot, you've made progress but the business is not a rocket ship. You're in the "ugly adolescent stage" and not sure how to get out.
Hiring Growth & Culture - The founding team is tight-knit, fired-up and all have real stakes in the business. Now you're hiring employee #10 who will have a small (and unlikely to be life-changing) equity and hasn't shaped the product from Day 1. How do you keep a culture that is aligned and excited as you grow? How do you add some more process without hampering the energy?

Most startup blog posts and accelerators are focused on the early stages so for the obstacles of the middle game you need advisors/mentors who know your specific situation and know how to read the board.

Both Sides Get to Move

When strategizing in chess you're constantly thinking about what your opponent will do in response to your move (or series of moves). Certainly this applies to literal competitors (although the true competition for startups is indifference) but more importantly it applies to other situations. For example transparency is a motherhood-and-apple-pie value at this point but often overlooked is the practical implications. When everyone knows when the sales numbers are consistently falling short they will make plans. Or if debating a pivot that might be exactly the right move for the business, it may be less attractive to certain team members to work on a B2B app instead of a consumer app. Each move reveals information that you must assume others will respond to.

Related to this is a mistake made by a lot of novice players - hoping that their opponent makes a mistake. They will pray you don't notice their attempt to fork your queen or skewer a rook to your king. These weak gambles often lead to poor positions and wasted moves that could have been better spent. In startups an equivalent mistake is trying to short-change your new employee in their equity grant (or not revealing the percentage ownership) and then being surprised when they don't have enough of a stake to turn down other offers later. Hoping that your opponent makes a mistake is usually a poor strategy that only works against weak players.

Playing Informs Instincts

In chess the more you play the more you recognize patterns and opportunities at certain positions. Chess is a logical game but there are many decisions made because you feel a move is stronger than the alternatives without concrete proof. While you can chart out a line of attack there is no certainty that is how it will play out. The more that you have seen a seen a similar situation the better developed your instincts become about what it can lead to. Unfortunately it is easy to study the chess games of others while it is very difficult to study other's startups. I have loved reading books like Founders at Work but most of the time it is nearly impossible to know the real story with enough detail to meaningfully analyze the success or failure of a startup beyond the conventional wisdom. So having gone through the startup game a few times and having a trusted network of advisors/mentors/board members is invaluable.

Reverse the Startup Curse

2011-03-25T00:00:00-04:00

[This originally appeared as a guest post on Greenhorn Connect]

I was born and raised in the Boston area so the most recent dust up about Boston vs Silicon Valley has me comparing it to Beantown's previous long-lived case of self-flagellation - The Curse of the Bambino.

For the Red Sox it was the trading of Babe Ruth, while for Boston web startups, it was the loss of Zuckerberg and Facebook to Palo Alto. That sin has cast loads of self-doubt on whether Boston can ever truly compete, often with those in the Boston tech scene being the most critical of its shortcomings. Like Red Sox Nation living under The Curse, those in the Boston startup community can often feel like losers despite being in the 2nd (or perhaps 3rd) best market to start a company. Compare this to the confidence of those from Boulder who have the Moneyball-era Athletics' knowledge that they are punching above their weight for a city of 100,000.

The good news is that we're no longer in our Mo Vaughn / Mike Greenwell era of frequent disappointments. During my five years around the Boston startup community I've seen things improve considerably to where I think we're assembling our "2004" roster:

Like David Ortiz, Hubspot is swinging for the fences with their recent $32M raise, and Dharmesh "Big Papi" Shah has shown how to remain a hacker while dominating marketing and growing a company filled with talented folks.^[1]
Long-time leaders like Bill Warner have been the squad's Jason Varitek, teaching younger teammates how the game should be played.
SCVNGR and WHERE resemble a certain Manny Ramírez - not always well-understood but looking to hit it deep. SCVNGR's Seth Priebatsch keynoting SXSW and building LevelUp to challenge Groupon and LivingSocial has a certain air of "Manny Being Manny" to it.
Companies like Gemvara, CSN Stores and Viximo remain modest while consistently putting up numbers over .300 like my favorite unsung hero, Bill Mueller.
Folks like Greenhorn's Jason Evanish, Victoria Song and the DartBoston crew have that Kevin Millar "why not us?" attitude that sparks others to do more. (Just avoid the Bronson Arroyo cornrows, please.)
Lean startup devotees Performable and Blueleaf have the iteration speed of bearded Johnny Damon.
The gaming startup scene is showing Curt Schilling resilience, despite his own 38 Studio moving across the border to Rhode Island. While a tough year for some companies and frustrations at not extending a tax credit to game makers, there is increased focus on gaming companies and PAX East in Boston was the largest PAX to date.
More smart, strategic Theo Epstein brains like Eric Paley at Founder Collective, Lee Hower at NextView^[1], and Antonio Rodriquez. And I can only imagine investors like Bijan Sabet and David Skok must have a John Henry-sized yacht to sail while making deals.
Finally, the pipeline of talent coming through TechStars and MassChallenge in companies like Localytics, SocialSci and Locately is producing Jacoby Ellsbury-esque^[2] up-and-coming stars.

This is not to imply there aren't real improvements to be made — there most definitely are — but the trend line is heading in the right direction, even if we lose a Nomar along the way. We'll never change Boston's weather and many will remain immune to the areas charms that include bright folks galore, (entrepreneur-friendly) health care reform, gay marriage and the nation's best sports city. So let's take a lesson from the "Idiots of 2004" to set aside all that baggage and get ready to play ball.

[1] Disclosure: both Dharmesh and Lee are investors in oneforty, the startup where I work.
[2] Yes, Ellsbury didn't join Pawtucket until 2007. It's just an (perhaps over-extended) analogy.

Favorite Albums of 2010

2010-12-29T00:00:00-05:00

It's that time of year again when the music I've loved from the last year is rounded-up to be recorded for posterity. There is some faux-science involved using iTunes playcounts to keep me honest about what I actually listened to, and not what I think I should have.

In reviewing the list of albums that were most often in my headphones I was surprised how many there were. 2010 was a good year for the music I like. I put together a iTunes Ping playlist with some favorite tracks from the albums below.

Honorable Mentions: Both Girl Talk and Kid Cudi put out solid albums similar to previous releases. Albums from Massive Attack and Beach House created dense, enveloping sonic environments. The soundtrack to The Social Network by Trent Reznor and Atticus Ross is fantastic, and perfect music to sling code to.

10. Jónsi, Go - An album full of bright, driving songs that combined with his unique voice can be epic at times. My only regret is missing his live show which was consistently described as highly entertaining.

9. The National, High Violet - While it didn't worm its way into my head the way Boxer and Alligator did, songs like "Bloodbuzz Ohio" and "Conversation 16" are fantastic.

8. Sufjan Stevens, Age of Adz - It took me a while to get into this departure for him, but it steadily grew on me. Seeing him perform it live helped illustrate what he was up to, and that the album should be played big.

7. Broken Bells, Broken Bells - Feels almost like a new twist on The Shins, in a good way. Songs like "October" and "The High Road" remind me what I loved about them in the first place.

6. The Black Keys, Brothers - Amazingly strong bluesy rock from start to finish, with a few standouts like "Tighten Up" and "Next Girl".

5. Vampire Weekend, Contra - I think I liked this album more than most I know, but got hooked on "Taxi Cab", "Giving Up the Gun" and "Run".

4. Kanye West, My Beautiful Dark Twisted Fantasy - I almost don't want to like this album given his already over-inflated ego but it's full of hits like "Power", "Monster", "All the Lights" and a bunch of great tracks. "Lost in the World" is infectious and even "Runaway" with its ridiculous lyrics is catchy.

3. Local Natives, Gorilla Manor - Extremely well put-together release that's obvious on tracks like "Wide Eyes" and "Airplanes". My only critique is that when I saw them live the songs felt like they had a lot more energy and driving bass that is pulled back on the album.

2. Big Boi, Sir Lucious Left Foot - Complex, layered and beyond catchy hip-hop. Each song feels very different with disparate styles, beats and featured vocals that serve to show Big Boi's breadth of talent. I'm not sure I've heard something else that sounds like it.

1. Arcade Fire, The Suburbs - Love the grand themes and earnestness of these songs especially on tracks like "We Used to Wait" and "Modern Man". They have this mixture of nostalgia for the past and distrust of the present that resonances.

Molly

2010-12-28T00:00:00-05:00

Babies come with hats.

Making Twitter an Information Network

2010-11-30T00:00:00-05:00

Twitter is increasingly being described as an "information network" rather than a social network or "fun-like-ice-cream" novelty. That seems accurate, but the challenge is that Twitter is currently designed like a social network. Even with its innovative asymmetrical following relationships what you follow on Twitter are accounts ^[1] not "interests". As a result, it is difficult to consume information effectively and to tweet for disparate audiences. Adding capabilities for how Twitter surfaces the most interesting conversations would make it more valuable.

Separating Wheat From Chaff

The primary problem in trying to follow "interests" on Twitter is that signal-to-noise ratio of following people is often low, which increases the amount of manual filtering needed to find the gems. Twitter has made strides in suggesting appropriate people to follow which has certainly helped the initial user experience (and is a problem I've been invested in), but it falls short.

A concrete example: I use Twitter to track developments in Ruby on Rails and so follow a lot of Rails hackers. While they certainly tweet about Rails, many have digressions about the TSA, where they ate dinner, sports teams they follow, etc. Sometimes that leads to serendipitous discoveries, but most often I just want to know the state of the art in Rails without the distractions.

Some argue that the follow cost is low and that you can always unfollow someone. While that is true at the margin I think it misses the fact that there is an opportunity to improve the situation in aggregate. Those low-value (to me) tweets compete for attention and crowd-out other more valuable content. Based on sample data of Twitter users the average engaged user follows around 400 people ^[2]. Attention is a scarce resource that should be designed for.

There are several partial solutions to this problem today:

Proxlet - A clever app ^[3] that acts as a proxy to filter your stream. It can only show tweets with links, block annoying foursquare checkin notices, or mute hashtags during conferences. Its biggest limitation is only certain clients (twitter.com, TweetDeck, Tweetie) support alternate API endpoints.
Twitter Lists - Twitter's feature to group people can help in browsing a community or finding people but because they are mostly manually created and not content-based the lists are still noisy. (And few clients, including twitter.com, make Lists prominent relative to your stream.) For example, I track the Twitter API team but reading their team list it is rarely about the Twitter API.
Nutshellmail - One of several digest tools that will highlight those tweets that are RT'd and favorited most. A useful tool for those who are time-pressed but don't want to miss anything big.
TweetMeme - The most popular links on Twitter in high-level categories (Technology, Sports, Gaming, etc.). Browsing categories gives a good snapshot of what's hot, but isn't native to Twitter or personalized to my particular interests.

The Audience Problem

The other side of the same problem is not being able to target your audience when tweeting. This affects how people tweet. That sounds trivial but is important as Twitter is a gatekeeper to millions of conversations. Some strategies are to "tweet like no one is watching" ^[4], have a "professional" account or focus on an aspect of their persona. This presentation on the "Real Life Social Network" describes better than I could problems with having flat, homogenous network without context. For a quick example, look at someone's profile page and note the difference between their replies from their regular tweets.

A concrete example: I have a lot of interests -- technology, startups, music, photography, Boston stuff, etc. -- and a set of followers that are likely only interested in one or two topics, if at all. As I've had more smart people follow me I prefer not to bore them ^[5] with my random musings. So, I find I self-censor to manage my online identity in a way that I feel comfortable sharing with everyone from friends to investors.

If I Were Twitter...

There are gains to be made for the user experience and possibly for Twitter's monetization to more directly allow users to follow interests. Twitter is not a social network like Facebook, which is a good thing. Twitter has an opportunity to be an information network that directs relevant news to the right people in real-time.

I think there are many ways to approach this challenge but here are a few ideas to make the interest graph natively part of streams:

Give users more control over what content is shown from a particular user. There is currently a per-follow setting on whether their Retweets are seen. This should be expanded to include options to only shows tweets containing links or those in certain categories based on the contents of their tweets. (Perhaps "annotations" would be useful to have clients add meta-data about tweets?)
Marry "saved searches" with Lists to automatically create private lists to track niche interests. Only show the tweets from the list that match the topic (not literal text matching), and allow filters to restrict to content that is particularly popular or meets other criteria. Provide subscription capabilities that insert these tweets directly into user stream.
Suggest people and "interests" to follow based on the content of my tweets, not just the social graph. This would be more a "what to follow" similar to "who to follow" (and an improvement on the currently flawed "Browse Interests").

Related Reading

"The unbearable lameness of web 2.0" identifies several problems with current social networks. The "Auto-binning" idea is one I particularly agree with.
Brian Solis predicts 2011 will be when Twitter becomes a "full-fledged interest graph". Brian thinks: "Twitter’s monetization strategy lies in the ability to reach interest graphs, not social graphs".
"The Myth of Serendipity" on the difficulty of "social serendipity" and pre-judging what users want to see
Dave Winer on "What should Twitter's vision be?" includes interesting tidbits around being a news system.
John Battelle on "Twitter's Great Big Problem Is Its Massive Opportunity"
Jeff Miller illustrates "Twitter's Garbage Problem" and proposed filtering improvements.

Notes
[1] Of course accounts don't have to be people, they can be feeds, but you subscribe to all tweets from an account.
[2] Sample data from Twitter uses on oneforty. It excludes those who follow < 10 people to skip those who never engaged, and those who follow > 5000 people who are likely unable to read their stream.
[3] Proxlet is dangerously close to solving my "Rabois Problem", named for Keith Rabois. Keith is SF-based entrepreneur and angel investor who is at times very interesting, but he is also a Yankees fan. Being from Boston and a member of Red Sox Nation I'd like to scrub all pro-Yankees tweets from my stream (anti-Yankees tweets are encouraged). See also: Bijan Sabet
[4] For additional tips to become the most powerful Twitterer.
[5] This is likely all a reaction to having Chris Dixon briefly follow and then unfollow me.

Flowtown Acquires WhoShouldiFollow

2010-11-10T00:00:00-05:00

I'm pleased to share that Flowtown has acquired the technology behind WhoShouldiFollow, a Twitter friend recommendation engine that my friend and I created. Xconomy has a good writeup on today's news. Flowtown is a social marketing platform aimed at helping businesses engage with their customers. What is particularly exciting to me is that the WSIF engine can help Flowtown deliver actionable intelligence for how businesses can expand their network to listen to the right people and find new potential customers. Thanks to Ethan Bloch and Dan Martell ^[1] of Flowtown, both masters of the startup hustle ^[2] (and blog infographic) who have been great to work with during this process!

Wondering who to follow? http://whoshouldifollow.com/29 Apr 08 via webEvan Williams
ev

My friend Gary Elliott and I originally built WhoShouldiFollow^[3] back in early 2008 as a way to make Twitter more useful, and as an opportunity to experiment with collaborative filters on the social graph. We'd both been playing with Twitter since mid-2006 and had gotten past our "twitter is dumb" phase early (everyone has one) but were aware of how rocky it was to get started finding interesting people. Designing the service was a fun introduction to building on the Twitter API and presented interesting challenges to giving real-time, (hopefully) high-quality recommendations. More recently Twitter has rolled out a similar feature for twitter.com (with an API promised). While it may obviate some of the general usefulness I think there is ongoing value for businesses to discover the appropriate people to follow based on various criteria (influence, location, etc) beyond what Twitter itself exposes. More broadly, there is a revolution underway as businesses grapple with understanding what truly works using social media and engaging with the right people is at the center of that process.

As a final note, I'm bullish on hackers having side projects. WhoShouldiFollow was a side project for us and lead to a whole bunch of good things: learning new technical areas, job offers and modest financial benefits. It was built before I took the plunge into startup life (which makes it more difficult to find time), but for hackers in jobs that may not be sufficiently challenging side projects are a great avenue to explore new domains. And there is huge benefit to owning something from concept to shipping. So, onto the next one!

[1] Dan Martell is also an angel investor in oneforty, the startup where I work.
[2] The startup hustle is best captured in song by Mr Rick Ross
[3] Yes, we realized after the fact that it should have been "whom". What can we say, we're better at coding than grammar. (And even Twitter did it!)

The Plumbing Revolution: Developers' Improving Toolbox

2010-06-15T00:00:00-04:00

The best advance in my life as a software developer recently is having to spend less time building (and re-building) plumbing. Much has rightly been made of the move to cloud computing and the virtualization of computing/storage, but in the last year the trend has continued up the stack to include all sorts of services. The benefit to developers is spending more time focused on your product, and less on the myriad of necessary-but-ancillary aspects of building a webapp.

In my thinking about this trend, I'm making two assumptions. The first is that the scarce resource is developer time, not money. If you have zero dollars to spend then you have to scrounge. The second is that the goal is to optimize for iteration speed, instead of other goals like education or maintaining complete control. In the past I used to (perversely) enjoy learning about sendmail, ipchains and makefiles; now having to wade into those type of issues is a distraction from improving our site.

A few examples of what I'm thinking about:

SendGrid - I hate the Yahoo! spam filter after how many emails my apps have sent that it has incorrectly flagged as spam. I just want emails to get into someone's inbox, not become an expert in SPF & DKIM. The day after switching to SendGrid from AuthSMTP this wasn't a problem anymore. Love it.
Chargify / Recurly - Billing and subscription billing in particular is a huge pain. You just want to get paid but have to worry about merchant accounts, vaults, payment gateways, PCI compliance, etc. Oh, joy. Several companies are trying to simplify this process for developers (which Sachin covers well in his Intro to Online Payments post).
Mogotest - the bane of web developer's lives is testing all the permutations of browsers & platforms that exist today. Firing up VMWare and manually testing pages is painful answer to know if your site works everywhere. We've been using the Mogotest beta and it has saved us hours of development time, and found things we would have otherwise missed. There is a direct correlation between fewer hours spent testing Internet Explorer and developer happiness. While this might not seem like plumbing exactly, having to ensure browser compatibility is a painful tax on creativity.
Apigee - the web has seen an explosion of RESTful APIs in the last ~5 years. Many are simple to consume with tools like ActiveResource but anyone who built on other's APIs knows you can spend a lot of time dealing with error conditions, latency, analytics, downtime, etc. Apigee is building a way to consume APIs by using their smart pipe between you and the the 3rd party.
Many more than I can cover: SimpleGeo for geodata storage & discovery, twilio for voice/sms-enabled apps, Zencoder for video encoding, Hunch for a recommendation service or BigDoor for adding game mechanics.

And it is easy to see why: developer's time is expensive and limited. The biggest danger for developers choosing to rely on other's platforms is that abstractions leak. Suddenly what had been a simple blackbox requires understanding the inner workings of these domains if something goes wrong or you outgrow the service. Still, for startups it is a smart move to trade some money to buy more time to focus on your core business and customers.

I think there are a number of areas ripe for significant improvements in infrastructure and how developers spend their time:

SEO - Having developers spend much time attempting to deeply understand Google is a net waste. And yet, Google is how normal people find things on the web, so it is too important to ignore. And the mantra of just "build good content" can be simplistic in practice. It is a difficult problem to solve but there is some recent activity in this space with companies like BrightEdge.
Understanding user behavior - Collecting actionable data on your users is challenging. Who are your most engaged and what motivates them? What was missing for those that visited and never returned? Tools like Google Analytics aren't auditable, and feedback services like UserVoice may miss apathetic visitors. There are several startups attacking various aspects of the problem such as KISSinsights gathering better surveys (Hiten is the man in this area), Performable A/B testing behavior on landing pages or UserMojo trying to analyze user emotion.
Debugging and responding to issues - One of the most difficult aspects to web programming is the number of moving parts. While it is a good goal to get to a state of zero errors there is often a steady mix of real issues (ones affecting real people), and those due to strange circumstances (badly formed requests, bots). We use Hoptoad and New Relic that are both valuable but neither completely solve the issue of discerning signal from noise. It can challenging to quickly know if a real problem that requires attention immediately, or if it is just background noise. Services like these could monitor constantly but only notify when levels have changed significantly. In addition to diagnosis based on data trends, it would be great to react to future exceptional conditions automatically (perhaps notifying subsequent users).
Hiring - While not plumbing per se the recruiting & hiring process can be a huge time sink for small development teams. While it is enormously important to build a great team it can be a serious distraction from customer learning and building the product. The existing models of recruiting and typical job boards both fail to provide enough screening for employers nor sufficient information for job seekers. Especially when you consider that most of the people that your company wants to hire are currently employed elsewhere.

Banksy in Cambridge

2010-06-03T00:00:00-04:00

Since the Banksy work across from my office on Essex Street in Cambridge came & went so quickly I thought I would preserve my snapshot at least.

Rolling Rails Deploys with Passenger

2010-05-31T00:00:00-04:00

We recently run into an issue where restarting Rails within Passenger was starting to take longer and longer as the number of plugins and gems we use have grown. This led to 30-90 seconds where the site was unavailable as Passenger restarted the Rails application spawner across every machine. That's far too long if you like to deploy frequently, which we do.

If you can, the best answer might be to move to Unicorn, the latest in a long line of Rails deployment options. It handles the process of migrating requests from old workers to new workers transparently. Awesome.

Our first thought was that Rails was just loading too slowly. Both Robby and I spent some time profiling Rails boot time and there wasn't a single culprit. Instead there were many contributors from the 50+ gems and 30 plugins we use that made it difficult to radically improve.

Next we took a different approach to have capistrano remove instances from the HAProxy pool and serially restart Passenger instances. To get this working there are a couple steps that I hadn't seen documented elsewhere (hat tip to Matt Conway's rubber for the serial task trick):

Change haproxy.cfg to perform a file-based health check for each backend: option httpchk GET /haproxy.txt In the rails app make sure that same file lives in the public directory so it will respond 200 in the normal case. Also, I needed to make nginx return a 404 when that file does not exist, since that is what HAProxy looks for to remove it from the pool. So in the nginx config:
```
if (!-f $document_root/haproxy.txt) {
    return 404;
} 
```

Change the default "deploy:restart" capistrano task to do the restart serially. Capistrano usually executes the task on all matching hosts concurrently so it requires a little hack to force it to run serially.

desc "Restart Passenger serially"
task :restart, :roles => :web do 
  haproxy_health_file = "#{current_path}/public/haproxy.txt"

  # Restart each passenger host serially
  self.roles[:web].each do |host|
    # 1. Take it out of the haproxy pool
    run "rm #{haproxy_health_file}", :hosts => host
    sleep(5)
    # 2. Restart passenger
    run "touch #{current_path}/tmp/restart.txt", :hosts => host
    # 3. Ping passenger to get it to warm up
    run "curl -s 'http://localhost:81/' &> /dev/null; exit 0", :hosts => host
    # 4. Re-add app to haproxy pool
    run "touch #{haproxy_health_file}", :hosts => host
  end
end

During deploys notice that haproxy will change the dashboard display to remove that instance and send requests to other instances. Be aware that this does mean not all instances have the same code running at the exact same instant. So for database migrations or other scenarios it might require putting up a maintenance page or coding defensively.

Downtime Postmortem

2010-02-08T00:00:00-05:00

Since joining oneforty last summer lots of things have gone well, but the mistakes we've made are usually more educational. The following is an attempt to capture the events that led to a brief site outage and some lessons learned.

A few weeks ago we rolled out an alpha version of our ecommerce platform and the news was covered on a few blogs, including TechCrunch. At roughly the same time (it seemed) there were alerts about the amount of swap space on one or more of our servers. The alerts would typically flap between a warning and then return to normal levels. I figured the two events were related and that the alerts were due to increased traffic, but not a serious issue.

Later in the evening as the alerts continued I investigated the situation. The site is built on Rails, running in Passenger and hosted on Engine Yard's EC2-based cloud service. Running passenger-memory-stats on our "application master" instance showed that there were about twice as many Rails processes as there should be, and there was a discrepancy between what passenger-memory-stats showed (total rails processes) and what passenger-status revealed (those that Passenger is actively using). There was less than 15MB of free memory and little swap left due to the stale processes. Not good.

Then I put on the straw that broke the camel's back. While trying to kill one of the stale processes, the machine locked up when it ran out of swap space. The Engine Yard configuration has the "app master" server double as both an application server and the load balancer, through haproxy, to the other application instances. This means that when that instance became unresponsive, the whole site went down. So now the clock is ticking (and I'm swearing to myself).

Engine Yard's service noticed within 60 seconds that the app master was unresponsive. It automatically killed the existing app master instance, promoted one of the other app clones to be the master and created a fresh app instance to replace the clone. This worked smoothly, except for two issues. When a new instance is created it is added to the load balancer before our gem installation is run, so there is a window of time when it would throw 500 errors. The EY flow of specifying required gems is through their web interface, instead of in our application's git repository. This is less than ideal (and it appears it might change soon), and we hadn't yet invested in a better workaround. Not wanting to wait for the gems to be installed, I terminated the newly booted clone.

Once the new app master was promoted, the site was back alive. The second problem was that EY doesn't automatically update the memcached config on each app server when an instance is terminated (also a known issue), so we were suffering increased cache misses that made the site very slow. I fixed the memcached config issue manually and the site was back to full functionality. Total damage was about 10 of downtime, and another 10 minutes of slow-to-unusable site performance.

Lessons Learned

I'm a fan of the idea of proportional investment when reacting to problems like this. The first instinct of most engineers, myself included, is that we need to build a sophisticated monitoring system, remove all single points of failure and have the site failover to redundant systems. Those are good goals, and maybe you eventually get there, but not until that level of investment is truly called for. Instead, we've taken the following steps:

Signed up for a more robust uptime monitor, Pingdom, for better email/sms alerts.
Fixed the issues causing stale processes. Initially it wasn't clear what was causing them to hang around after a deploy. The first step was to write a quick capistrano task that would kill any detected during deploys. This at least addressed the symptom. After more research (and a helpful pointer from EY's Ezra) it became clear that it was because of an interaction between A/B testing framework vanity's handling of redis connections and Passenger's forking model. A patch to vanity forced it to stop accidentally sharing a redis connection between processes to fix the underlying problem. (Passenger's model has real advantages but alters the "shared nothing" assumption many components make.)
Working to get to a point where alerts and notifications do not become background noise. When they do it is too easy to ignore them and miss real issues. I think this always sounds easier than it is. Webapps have a lot of moving parts and receive many odd requests that can trigger alerts from machines, exception trackers and performance monitoring tools. There will be ongoing work to find the right thresholds and to address issues as they crop up.

Favorite Albums of 2009

2010-01-03T00:00:00-05:00

Keeping alive my annual tradition (albeit a bit late) I've rounded up what I listened to in the past year. The list is mostly based on iTunes playcounts, with a little bit of editorial judgement. This year it seems like there was an order of magnitude between each position in terms of how much I liked & listened to these albums.

Honorable Mentions: The Dark was the Night compilation has some great songs/covers by My Brightest Diamond, The National and Bon Iver. Speaking of Bon Iver, his Blood Bank EP was solid and I really like "Woods", vocal effects and all.

10. Dinosaur Jr, Farm - It has some great guitar-driven rock, with standout track "See You". (Bonus points for reminding of being in 8th grade, which was a good year for discovering new music.)

9. Discovery, LP - A quirky album, that is heavily produced and at times too sugary, but pretty engaging on tracks like "Orange Shirt" and "Osaka Loop Line". It also found its way into my hands at the right time for some late-summer pop.

8. Isis, Wavering Radiant - Haunting and full of great dynamics, starting with the album opener "Hall of the Dead". For me, the vocals (both screaming and singing) largely detract from it, and don't feel like the right fit. I'd be curious to hear it as an instrumental album.

7. Tegan & Sara, Sainthood - Doesn't live up to The Con, but is still well-made, fun power-pop. I think they work best when their songs are more driving, like on "Don't Rush", "Hell", and "Northshore" where there is less time to slow down and get a bit overwrought.

6. Neko Case, Middle Cyclone - I haven't listened to much Neko Case, so this was a welcome invitation for me to plump her catalog. Her voice is amazing, and shines on songs like "I'm an Animal".

5. Propagandhi, Supporting Caste - I don't listen to a lot of heavy stuff these days, but I found myself reaching for this all the time. I was a fan of "Less Talk, More Rock" back in the day, but they are just musically more interesting now, and still just as infectious. And songs like "Dear Coach's Corner" still combine their style of leftist politics and shared Canadian bonds over love of hockey.

4. Passion Pit, Manners - I think I have to admit that I'm no longer immediately put off by electro-pop with often falsetto vocals. Songs like "Little Secrets" and "The Reeling" make it more impressive that this is their first full-length album.

3. The Thermals, Now We Can See - I'm a sucker for their pop-punk sound and lyrical themes. Less political than "The Body, The Blood, The Machine", it remains full of energy on songs like "When I Was Afraid" and "You Dissolve".

2. Dirty Projectors, Bitte Orca - An experimental album that got stuck in my head for weeks. Songs like "Useful Chamber" are doing a lot at once and meander around but don't lose me (especially with a fantastic build in the final minute).

1. Phoenix, Wolfgang Amadeus Phoenix - Great from the opening of "Lisztomania" to the close of "Armistice", and with more layers than I originally thought from hearing "1901" on the radio. Also prompted me to pick up It's Never Been Like That which is nearly as good.

Mac OS X Setup

2009-07-27T00:00:00-04:00

I just did the whole setup-a-new-Mac dance and thought I'd crash a faux-installment of The Setup to share my notes. My current machine is a 15" MacBookPro with 4GB RAM purchased two weeks ago. Overall a fast, shiny new machine. Which, coincidently, is my biggest complaint. The glossy screen is sooo reflexive that it is often like staring at myself in a mirror. (Why do they have a matte option only on the 17" MBPs?) The feel of the keyboard and battery life are a nice step up from my previous Apple laptop.

Most of my days are spent inside TextMate. I was an Emacs junkie with a giant elisp configuration, but found the corded key combos took a toll on my hands and I left it for Eclipse (back when writing Java). TextMate is a solid editor, with reasonable defaults and similar enough key combos for common tasks to keep me happy. I use a customized version of the Sunburst theme. I like ProjectPlus as a replacement drawer for better sorting options. I don't extensively use TM bundles, but found the Copy as RTF bundle helpful for making slides. The other place I spend a lot of time is Terminal. It is passable, and better after changing it to send Ctrl-H as Backspace and turning off the audible bell.

I like MailPlane for work email, Google Notifier for personal email, Adium for IM'ing, Tweetie for Twitter, Colloquy for IRC, NetNewsWire for password protected RSS feeds, and have been trying Google Quick Search Box as a launcher.

Music helps the code flow so I use the shareware Synergy to control iTunes from global keyboard shortcuts without leaving what I'm working on. Mojo is great to share music with friends. I prefer listening to albums more than random mixes so I customize the iTunes view options to include Track Number and Last Played (which are inherited by later created playlists) for easier sorting.

I primarily browse with FireFox, but end up using Safari and Opera (and now Chromium) as well. I'm most comfortable with FireFox but the memory leakage is frustrating enough to tempt me to switch. Often I find that videos are smoother to watch in Safari, as well as feeling snappier. I use FireBug, YSlow and Google's Page Speed for measuring web performance, ColorZilla to get the hex for colors and 1Password helps keep my passwords slightly less insecure. Also, I really enjoy the bookmarklets for Instapaper to read something later, and Readability for clean and enjoyable web reading.

The customization steps that I always have to dig up again are:

Stop the Dock from growing and shrinking. Then get rid of the drop shadows via:
defaults write com.apple.dock no-glass -boolean YES && killall Dock
Re-map Caps Lock to Cmd in the Keyboard Modifier Keys preferences. I never need Caps Lock and it makes a lot of shortcuts easier for my hands to type.
Switch the keyboard shortcuts to enable "Full Keyboard access" for "All Controls" so pressing Tab will land on Submit buttons. As a keyboard fan this is a must.