Philly Emerging Tech – Day 2

Today was the second day of the 2009 Philly Emerging Tech conference.

I was slightly worried about the traffic, so I left nice and early to ensure I’d be there on time today. Fortunately, the weather and trafic were good and I actually knew where I was headed so I arrived with plenty of time to spare. With about 45 minutes until the first talk, I grabbed some fruit and sat down at a large table table to figure out which presentations I was going to attend.

A short while later a towering fella with a moustache came and sat down next to me. I recognized him from the day before. At about 6′ 6″ and the only white guy with a stash, he’s hard to miss. Jeff, it turns out, works for a local company that creates mapping software and he was at the conference to pick up new techniques that he could take back to apply to their processes.

Maybe it was just a coincidence or maybe it was just because he was a giant and I kept noticing him when he was near me, but it seemed like Jeff was never more than 10 feet from me throughout the entire day. It became this awkward running joke because every time I went to the restroom he would somehow wind up being next to me in line. Anyway…


Andy Hunt was up first.

When I read books, I highlight the stuff that’s meaningful to me. If, after a few chapters, I haven’t highlighted anything, I usually put the book down and find something else. Occasionally, the opposite is true. There will be so many good lines that I have to stop highlighting, because it defeat’s the purpose. 37Signal’s Getting Real comes to mind.

Andy’s presentation was like that. There were just so many good, practical take aways.

Here’s a sampling of the topics he talked about, which were derived from his recent book Pragmatic Thinking & Learning: Refactoring Your Wetware.

Experts have the ability to recognize patterns in a given context. For example, doctors can sometimes diagnose a patient’s illness from a quick, cursory glance without knowing exactly what led him or her to that specific conclusion.

It used to be a commonly held belief that you could not grow new neurons. You could kill them, but you couldn’t grow more. The problem was that they were basing that conclusion on tests performed on primates held in captivity. If you exposed the same primate to an environment filled with rich sensory inputs, you’d  discover that it was possible to grow new neurons.

The Dreyfus Model of skill acquisition represent’s someone’s journey from notice to expert. An expert’s relationship to the system changes. Experts rely more on an intuition and an internalized body of knowledge instad of rules. They had expert airline pilots document how they made their decisions. When given to the beginners, it improved their performance but when they asked the same experts to follow those rules, it degraded their performance. In the Dreyfus Model, novices are detached observers of the system. Experts, on the other hand, are a part of the system.

If you criticize the system, the expert is more likely to take offense to it.

How do you become an expert? Delibrate practice. Imitate, assimilate, innovate. First you following along, then you make it part of your experience, then you change it. It takes about 10 years to master a subject. He gave Linus Torvald and Motzart as examples. If you imitate a smart person, it actually makes you smarter.

Note: I think Mr Hunt reads Malcom Gladwell :)

Scientists used to thinking that the right half of the brain was used for creative, innovative activities and that the left was primarily used for linear functions. It’s actually a bit more complicated than that.

Linear skills are things that we computer science guys know a lot about. It’s the logical, analytical skills that we use everyday to do our jobs well. Nonlinear skills, however, are the hallmark of the expert. It’s not rational. It aggregates a whole bunch of things together. It’s more about synthesis than analysis.

You process aesthetically pleasing things better.

Saying that your iPod can store 5 gigabytes is a lot different than saying it can store 20K songs. Grandma doesn’t know.

This mad psychologist scientist named Lozonov did an experiment where he had his students relax and listen to hippie music before learning new material. When compared to a control group, they had much better retention.

Did you know… Thomas Edison would take naps with ball bearings in his hand before so that when he finally dozed off they would fall and wake him up. He would then immediately write down on a notepad whatever he was thinking at that instant.

Mediate. It can do amazing things.

People have lots of cognitive biases. We’re not rational beings. Our consciousness gets inputs from our memory, from our sensory inputs, and from our imagination. Our mind is not so good at discerning which though.

The relationships between things are often more important than the things themselves.

Think about the color red and all the sudden you start seeing it everywhere.

When you read, one method to improve comprehension is to follow the SQ3R method: Survey, Question, Read, Recall, Review.

Bobsledders visualize the course before they begin. It helps them perform better when it’s game time.

Multitaskng, as in constantly checking your email, robs you of 20-40% of your productivity.

Context switching, like switching applications on a computer, is a cognitive train wreck.

Organizing applications by the task helps you avoid context switching.

Something about a pigeon banging its head against a tree.

… I thought his presentation was the most interseting out of the 15 or so I saw the last two days. That book is on its way.


Next Up: Mike Culver, who gave a great introduction to cloud computing and Amazon’s Simple Storage Service (S3). I didn’t know much about cloud computing beforehand, but Mike made sure to remedy that.

We need more innovative companies like Amazon.


After that I attended a talk on Scrum by Jim York. Scrum, from what I gathered, is an agile software development technique that focuses on breaking up objectives into one month groupings called sprints. He had a lot of good lines too:

“To be successful, an organization must produce something of value.”

“How do you create value?”

“Organizations create value when they deliver to the customer:

1) What they want

2) When they want it

3) At a price they’re willing to pay.”

“When task is made king:

1) Management creates organizational structures best suited to accomplish the tasks

2) Efficiciency experts establish the best way to execute the tasks

3) Quality experts implement controls to ensure processes stay within defined limits

4) Workers are trained in processes to complete the tasks

5) Management evaluates performance on process compliance”

“Where is the innovation?”

Management and Quality Assurance, NOT the ones actually doing the work.

Change disrupts process. Most processes are change resistant.

Work is prioritized from the customer’s perspectuve.

Customer is king.

Identify your ultimate customer.

Involve the customer.


The third talk was by Jascha Franklin-Hodge, who helped cofound Blue State Digital, the company that led Obama’s tech campaign. Few notes:

Blue State Digital was cofounded in 2004 by four former Howard Dean staffers, one of whom was Jascha.

The company currently has more than 75 employees and 200 clients.

There’s three big parts to what they do: design, which involves incorporating your vision of what you want the users to do; technology, such as the donation process; and strategy, which is tackling how you want to achieve your goals… “Not just taking brochures and putting them online.”

1B emails to 13M addresses


200K offline events planned over the web

35K local volunteer groups

14.5M YouTube minutes for official videos

The Obama campaign raised $770M dollars and more than half a billion of that was online.

He then went over several overarching themes for how they did this:

Drive Action – the Obama homepage was “all about what you can do”. Each medium has its own set of activities. iPhone app would sort your contacts by area code, listing those in key battleground states at the top of the list. 70% of actions come from the top 10% of people.

Be Authentic – emails from people (Barack, Michelle, David Plouffe), not from namesless org accounts. Notice how Google blog posts are by the engineers who developed the product.

Turn Users into Advocates – recognize your leaders and engage them. Invite people to weign in. Collect user input and reflect back on the good stuff. Solicit ideas from the community and actually use them. Connect people with each other.

Be Relevant – Palin’s speech resulted in $11M being donated — #1 contributor. They took advantage of the situation.

Build a Strong, Open Brand – Make it professional. Same logo was used everywhere.

Measure Everything – “If you’re not measuring it, if you’re not testing, you don’t know what you’re missing.”

They did A/B testing on donation forms to see what people responded to the most.

Not a lot of tech savy people in the political world


After that I attended one more Scrum presentation, this time presented by Mike Vizdos. For some reason, I couldn’t help but think of Crum from Ahh Real Monsters whenever anybody said Scrum. Come on… “Scrum Master”? IMHO, they should have picked a better name.


And last, but certainly not least, was Andy Hunt again. Few more noncoherent notes:

“Without excellent personnel, even good to excellent proceeses can only achieve marginal results” – Casper Jones

Talent matters. People make a big difference.

The real world is messy.

Kaizen – Feedback and continuous improvement. Have a short feedback gap.

Systems Thinking – Problems with any person on the team are problems for the entire team; it’s all one system.

Risk management – minimilism is a good approach for dealing with risk.

ROI – make your product useful sonner rather than later.

Continuousness – Don’t take too big of a bite at any one point in time.

((((do something!) small) useful) now!) – Bob Berner, father of ASCII

Feedback – “Agile development uses feedback to make constant adjustments in a highly collaborative environment.” – Practices of an Agile Developer

Ask yourself periodically:

1) Why am I doing this?

2) Does it have to be done this way?

3) Does it have to be done at all?


On a final note, as I was leaving the parking garage there was Jeff again.

He told me to get my belts replaced. =)

Philly Emerging Tech – Day 1

Today was the first day of the annual Philly Emerging Tech conference. Here’s a quick rundown:

I left about an hour and a half early to allow plenty of time for Philly traffic but despite my GPS’s estimate of a 40 minute commute, I wound up arriving 15 minutes late. The bad weather and a wrong turn into Camden didn’t help anything either.

Today’s keynote speaker was Michael Tiermann, Vice President of Open Source Affairs at Red Hat and President of the Open Source Initiative. To give you an idea of his vision, here are a few quotes from his presentation:

“Lots of innovation results in an increase in productivity”

“Now that we can do anything, what should we do?”, quoting Bruce Mau

“How can we be better?” quoting JP Sloan

“Leave your system open to innovation”

Did you know… Somebody did a study of contributions to Apache and calculated that 1 developer did about 20% of the work, 5 did about 50%, 15 about 80%, and an amazing 388 people to do all 100%? Also, proprietary software averages 20-30 defects/1000 lines of code. Open source: less than 1. The linux kernel is about 5M lines of code. An automated software scan came up with 985 errors and with the help of the community, they were all fixed within six months. Now compare that to Vista, which is estimated to be about 50M lines of code, which does not have an extensive community to help fix what must be at least a few hundred thousand lines of defective code.

He said something else that I thought was good. I forgot the context, but it was something like “The cost to the developer is less than the value to the customer.”

When he finished I went to an introductory presentation about iPhone software development by Bill Dudney. I never really appreciated how easy it is to create an application’s interface. I thought you had to program the behavior of the tables, the sliding buttons, etc. Turns out most are just customizable controls. He walked a packed room through the creation of a simple app in under 40 minutes. He was very well spoken and definitely knew his stuff.

I bounced around a bit during the next hour. I started off in a presentation about Android development then went to Exhibitionism in Software Development and finally wound up in a talk being given on the importance of accessibility in web development.

After that was lunch. I thought we would have to leave to go get lunch, so when I came out of the accessibility talk and a buffet was already set up, it was a pleasant surprise.

I ate with a few other people from the Philly on Rails meetups–Chris, Jon, Angel, and Randy. Colin, Alex, Aaron, JP, and a few others were around too. Also met Chris, the CTO of a Philly startup called Vuzit that has created a novel Ajax-based document viewer.

After lunch was a talk called Innovation in Ruby given by Jason Seifer and Gregg Pollack of Rails Envy fame. Their presentation was excellent both in terms of content as well as how they spoke and interacted with each other. For some reason I kept thinking “Batman and Robin” the whole time. Anyway, a lot of it was over my head, but I left with a much greater appreciation for the brilliant work being done in the Ruby and the Rails communities. I also briefly met Ezra Zygmuntowicz, who apparently founded Engine Yard and created merb. Nice guy.

Next was John Resig of jQuery glory. Before the talk I asked the guys why use jQuery over Prototype. I don’t remember what Randy said, but it was something poetic about how code just flows from his hands or something to that effect. John’s talk was good, despite the Public Address problems that resulted in us hearing the presentation being given in another room and eventually a full blown rock song. Next project is going to be with jQuery. I’m convinced that it kicks ass.

Last but not least was Mike Culver from Amazon Web Services who spoke about and demoed Mechanical Turk. I thought his presentation was the most interesting one all day. What an amazing technology.

Tomorrow: Day 2, where I continue to learn more about just how much I don’t know. :)

Back to Work

Somehow about two weeks went by without a update to Domain Pigeon. Domains were still being released (that’s all on autopilot now) but I hadn’t made any changes to the site. I didn’t think it was that long, but when I started working on some new features tonight I realized just how long it had been since I last opened TextMate or ran a script/console command.

As is normally the case, the time off resulted in some innovative new ideas for the site. With those in mind, I’m refreshed and ready to start working again.

On an unrelated note (maybe), I came across an excellent Washington Post piece titled ‘Pearls Before Breakfast‘ which is much needed reminder to slow down and enjoy to the ride.


Not much to report on.

I spent a few hours this week at a local sandwich shop reading Beginning Python and my initial impression is extremely positive. My next project will likely be made with Python and Django.

I’m also working on some potentially big updates for Domain Pigeon, which I hope to roll out in a week or two. I realized that I’m in a great spot to just experiment with the site and see what happens. If the changes work, great, if not, no big deal. After all, failures are excellent opportunities to learn what not to do which can often be just as valuable than knowing to do. :)