NOTE: This blog has been moved to

Sunday, December 16, 2012

Indispensable Principles for Debugging - Book Recommendation

Don't judge a book by it's cover (or it's website... or it's poster). Especially true in this situation. David Agan's book "Debugging: The 9 Indispensable Rules for Finding Even the Most Elusive Software and Hardware Problems" is a great find, chocked full of simple wisdom on how to find and fix bugs. Some of you will be turned off by his use of Comic Sans (see his note at the bottom of his website for his funny comment on why he sticks by his comic sans guns).

What it is not.
It is not a tutorial for any debugging tool. It is not a primer for debugging in any specific language. It is not an exhaustive tome on debugging. It doesn't have specifics on debugging for the web.

What it is.
Short and sweet. It contains 9 principles that are timeless and can be applied to any system you are trying to debug. Timeless aspect.

Why this book?
I have often wondered why some engineers seem to be able to debug and fix issues and others just create a mess when they try. Of course some minds work in a more systematic manner than others and some have more insight. This book teases out the principles that underly finding and fixing issues. The principles are dead simple and completely obvious when you glance at them. But I promise you that a lot of our colleagues don't practice even a portion of these.

I also like this book because it is a little older and it doesn't try to tie itself to the current trends. One of his first "war stories" is about debugging one of the original Pong games. Love it. First published in 2002, and then again in 2006, it is kind of hard to find. I bought the last 7 copies from ebay today but you can get it from the kindle bookstore.

Nine Principles

  1. Understand the system
  2. Make it fail
  3. Quit thinking and look
  4. Divide and conquer
  5. Change one thing at a time
  6. Keep an audit trail
  7. Check the plug
  8. Get a fresh view
  9. If you didn't fix it, it ain't fixed

Do you have other suggestions for teaching the basics of debugging? Comment below.

Friday, November 23, 2012

User Interface Engineering Disciplines & Skills

One of the first things I did when I joined PayPal was rename the "webdev" to "User Interface Engineering". Now while there is nothing wrong with the title web developer, at PayPal webdevs were considered one step below a "dev".

Changing a title in itself is fairly meaningless. But at the same time names are powerful. Over the last year I believe we have been really growing into the title. Recently I started sketching all of the engineering disciplines as well as web development skills needed for this role. Here is the diagram I created.

Engineering disciplines & skills needed for UIEs

Here is an accessible version of the diagram above.

One of our next steps is to flesh out curricula that ties to the diagram above. We are calling this courseware "Bring Design to Life University" since our job as UIEs is to bring great design to life.

List of Mockup/Prototyping Tools

Started keeping a list of these tools. No particular order.


Axure RP



Balsamiq Mockups































GUI Machine

inPreso Screens









Added after original publishing of this article:


Twitter Bootstrap


Saturday, October 06, 2012

Lean UX & Agile Development. Rationalizing the Two Methodologies

I really enjoy Silicon Valley Code Camp. Lots of speakers, lots of local attendees. Very informal and always a good crowd. No different today. Really engaged audience.

One topic I covered was how Lean UX and Agile work together in harmony. Here is a diagram I shared that I have been using at work to explain how they are related and how they are different.

Lean UX & Agile: Two separate but aligned workstreams
Agile focuses on iterative delivery. It is by nature engineering-centric (though it has been applied to non-engineering tasks) as it produces software meeting specific acceptance criteria. It has numerous ceremonies instead of heavy process like waterfall methodologies require.

Lean UX focuses on refining the experience through collaboration, shared understanding and continuous customer feedback. While agile finds its roots in engineering, Lean UX stems from work borrowed heavily from Eric Ries in the Lean Startup and applies it to refining customer experience. Lean dispenses with agile ceremonies but like agile also dispenses with extraneous documentation. Lean UX gets its cadence from getting prototypes or mockups in front of customers regularly while agile gets its cadence from delivering working software released at the end of each sprint.

In the above diagram I call out this separate stream of work -- the Lean UX stream. It is separate from the agile streams, but does not replace the agile streams. It runs ahead of the agile streams like Sprint 0 methodology but it does not stop with sprint 1 (nor is it confined to starting one sprint ahead). It continues to run in parallel with the agile tracks. What gets learned in the LeanUX stream (essentially a continuously evolving living prototype) gets fed into the agile streams and becomes some of the user stories. Its cadence may change (we usually start with weekly usability testing and then slow down to every 2 weeks or so) but its purpose does not. Separately I have noted the technology changes we have introduced that allow us to have seamless movement between Lean UX streams and Agile streams with respect to software delivered. This is an essential ingredient to making this work well.

Lean UX & Agile work together well. And while they share lots of common principles they are distinct work streams. Lean UX & Agile aren't the whole picture on how to create great products. You also need to be able to gather customer insights, define customer problems, refine solution concepts, deliver & test solutions and then rinse & repeat these steps. Lean UX is one tool to help us refine & test solution concepts. Agile helps us deliver solutions. A/B testing helps us test what we deliver. Customer driven innovation (pioneered at Intuit) defines a broader methodology that Lean & Agile fit nicely inside of.

Ok, so with that context here are the updated slides from the talk I gave earlier today on Anti-Patterns for Lean UX. You can see an earlier version on slideshare.

Are you using Agile & Lean UX methodologies? How are you approaching innovation & experience delivery in your organization?

Saturday, September 08, 2012

Three Key Principles to Operating as a LeanUX Team

Coaching teams here at PayPal on what it takes to operate like in a LeanUX manner. I have typically boiled it down to just 3 key principles.

Three Principles

1. Maintain a Shared Understanding at All Times

This was key when I was at Netflix and I see it even stronger now. A lack of a shared understanding means you will have have to lots of documentation. Having documentation that is predictive in an iterative environment is problematic. It is not realistic since you need to be able to change assumptions and solutions early in the cycle. Historical documentation becomes more necessary as teams downstream need them.

A good way to alleviate much of the documentation normally needed is to pull the teams that are normally downstream (like QA, Localization QA, Localization, etc.) and bring them up further in the earlier cycles. Additionally you can create tools earlier in the process that proxy for these teams and reduce the number of those representatives you will need. Tools downstream should be automation driven as much as possible.

In a recent LeanUX project I kept referring to a/b testing. At some point it became apparent that not all the team understood what I meant by a/b testing (remember this is what I ate, drank & slept at Netflix). It was an easy solve to get us a shared understanding, but the mistake was assuming we had one already on all topics.

2. Collaborate, Collaborate, Collaborate

Walls between teams get erected whenever teams get big, become overly specialized and work in isolation. Agreeing to work as a team, trust each other and make collaboration the underpinning of the team's ethos will ensure the walls will come down.

It's not easy. We like to retreat back to the comfort of going solo or dictating to another the way things work. How can you tell where the walls are? It will be at those hand-off points that happen across disciplines. Collaboration is the opposite of this toss it over the wall mindset.

It's ok for teams to take time to work separately or have sessions to reflect. But if you are doing this to avoid working as a team then watch out. Trouble is already at your door.

3. Build/Test/Learn -- or Get to the Customer Early & Often

This is the secret to getting rid of team politics, getting rid of the prima donna, getting rid of analysis paralysis. Getting to the customer in usability tests, research, a/b testing and so on and doing it on a weekly or bi-weekly basis if at all possible keeps the team pure. It is the difference between a stagnant pond or a clear lake fed by streams. User data & user feedback will keep the team focused on the customer and the real goals of the product. 

Believe me this really works. It is astonishing to see how much ownership takes over the team when they sit in all the usability sessions as a team and watch the successes and failures of their work. I have seen the full team time & again go out of a session that didn't go so well with a renewed determination to nail it next time. Customer feedback is the elixer to what ails you (now I just have to mention that all of the standard practices of how to gather this feedback and what to do with it still apply).

It's Pretty Simple

Fragile is often the name given to agile projects that are busted. I have used the term "Lean-Fall" to describe that ugly mix of waterfall & LeanUX. When I talk with teams that are struggling to work in an agile manner it could be a host of issues. There is a lot of machinery from an agile practice standpoint that might need tweaking.

But frankly, over and over I have seen it come down to one or more violations of these three principles. If you don't have shared understanding, collaboration and continuous customer feedback, no amount of scrums, scrum of scrums, agile coaches, etc. is going to make a difference.

View what you are doing as a team in light of these principles and see if it doesn't give you a fresh perspective on acting like a startup.

If you like this article, you might want to check out my Anti-Patterns for LeanUX presentation.

Thursday, September 06, 2012

The Who, What, How of Satisfaction for Frontend Engineers

This is an over-simplified view. But I was thinking about what really matters to frontend engineers in their jobs (besides money).

It matters who they work with

  • They want to be challenged by people smarter than themselves.
  • They want to work for someone who gets it and values what they do.

It matters what they work on with who they work with

  • The experience must matter. It must not be an afterthought.
  • The technology stack must be industry standard and make development fast.
  • The work must be relevant to our customers.

It matters how they work on what they work on with who they work with

  • They want to partner early & often with product & design.
  • They want to work in a lean, collaborative environment.
  • They want to have customer feedback integrated throughout the lifecycle.
  • Some parts of their work must be given back to the open source community.
  • They must be valued for the unique contribution they make.
  • They must be able to grow in their job to reach new heights.

What do you think?

Wednesday, September 05, 2012

Forming a new Accessibility Team at PayPal -- Welcome Victor Tsaran & Srinivasu Chakravarthula!

I am excited to share that we are formalizing the already ongoing effort at PayPal in Accessibility by forming an official Accessibility team.

And to kick things off right, I am thrilled to announce that Victor Tsaran will be joining PayPal on Monday 9/10 to lead this team. Many of you know that for the past 7+ years Victor has been the face of Accessibility at Yahoo (along with the excellent Alan Brightman) And if you were reading my blog back in 2005 you will also know that I wrote a piece about Victor's impact on my views about accessibility. I was just 2 months into my career at Yahoo! Here is a snippet from that article:
Ok, I'll start with a confession.
I think accessibility issues have always been an abstract concept to me.
It usually was an afterthought, something that the usability folks dinged us for. You know the text wasn't dark enough or the font was too small. It seemed to me that every experience I had with accessibility was from the negative perspective.
You see, I love rich interfaces. I love visualizations. I enjoy pushing the envelope. Somehow in my mind I just saw accessibility and richness as mutually exclusive.
This all changed over the last two weeks. It happened almost the first time I met Victor Tsaran, Yahoo!'s Accessibility Evangelist/Manager. Victor is an incredibly bright engineer who happens to be non-sighted. If ever there was an evangelist and champion for accessibility, Victor is the man.
Victor does not come at accessibility from the negative aspect. Not at all. He approaches it with an enthusiasm, a sense of humor, and a challenge to create rich interfaces that are richly accessible.
My thinking hasn't changed. That's why when I started thinking about forming this team there was only one person in my mind -- Victor.

Here is a great intro to Victor and his work at Yahoo!

In addition, I want to call out an equally incredible hire by our QA team in Chennai, India -- Srinivasu Chakravarthula. Vasu was the Senior Manager, Inclusive Design at Yahoo! Bangalore. Vasu is well known in the industry (like Victor). You can follow Vasu on twitter @vasutweets.

We are stoked to have these two great advocates for accessibility on the PayPal team. We look forward to bringing Inclusive Design to the fore on all of PayPal's products & services.

They will join together with the excellent work that Cathy O'Connor, Dennis Lembree and Nawaz Khan have been doing to bring accessibility to the fore at PayPal to form this multi-disciplinary team (design, engineering & QA).

Welcome Victor & Vasu!

Tuesday, September 04, 2012

Looking for "UI Engineer, Developer Experience"

Just one of the many places we are upping the game at PayPal is in our developer experiences. A few months back PayPal formed an Application Platform team. Part of its charter is to create great developer experiences as well as simple developer APIs. Deepak Nadig, the Technical Director of this team, is looking for two UI Engineers to work on exciting projects in this area. Here is part of the job description:
We are creating a world where ‘payments’ is synonymous with PayPal. The PayPal Application Platform is a suite of web services, which are used by internal and external developers to enable payments anytime, anywhere and anyway. The Application Platform is complemented by a comprehensive Developer Experience that enables these developers to find, learn and use these APIs effectively. The Application Platform processes billions of requests each month, and enables PayPal and its partners to rapidly innovate on new payment scenarios and enable new experiences. Therefore, the evolution of the Application Platform is integral to PayPal’s long-term strategy. 
As a UI engineer in our development experience team you will be responsible for the development and delivery of the web experience used by developers to find, learn and use our APIs. You will work closely with our product and experience and/or integration teams to understand the developer needs and deliver the experience that meets their needs.
 Please reach out to me directly if you are interested. bill.scott _AT_

Monday, September 03, 2012

Why You Should Work with Me at PayPal

I have blogged a little in the past as to why I came to PayPal. Today I want to talk about why you should come to PayPal to work with me.

To start with let's be frank. Last year when I contemplated joining PayPal it was clear PayPal had long lost its luster as the darling innovator of Silicon Valley. PayPal had changed the face of payments when it came on the scene. And the innovators DNA is evidenced by one of its founders going on to create the first viable electric car and the successor to the Space Shuttle program. To get a fuller look at the innovator DNA of early PayPal, check out the wikipedia article on the PayPal Mafia.

But what I sensed when I joined PayPal last October was a desire to get back to its innovation roots. Back to a desire to innovate again. This desire coupled with a very healthy financial position seemed to me to be a no brainer. I decided to take the chance. So what has happened in the past 10 months? From my perspective three things: New leadership, new way of working and new talent.

New Leadership

How often do companies the size and age of PayPal get a complete makeover? Not often. But that is just what has been underway here. Take a look at just some of the new leadership in the company:

  • David Marcus. President. David came into PayPal a year ago as part of the Zong (mobile payments) acquisition. David is a product, user experience, social & mobile guy at heart. He is passionate about innovation and his motto is GSD (Get sh*t done!). Before PayPal I believe the largest organization he led was 200 people (Zong). He believes in small teams empowered with really smart people working with great business context. I could not be happier than I am working at PayPal with David at the helm.
  • James Barrese. James became the CTO of PayPal earlier this year. James brings the challenge to think differently to the technology organization. James also brings a wealth of experience leading technology at eBay with lessons learned from what worked and what didn't. 
  • Hill Ferguson. Head of Product & Mobile. Hill also joined PayPal through the Zong acquisition. Hill has strong mobile & financial roots and is a passionate leader bringing together all of our product & mobile teams together in a single team.
  • Hendrik Kleinsmiede. VP of Global Design (UED). Hendrik joined in January and brings with him a passion to create great experiences by attracting top talent and collaborating deeply with product & engineering. Hendrik & I have a very close working relationship and in our short time have collaborated on bringing LeanUX and Mobile First thinking in as first class citizens of PayPal.
  • Kirsten Wolberg. VP Technical Business Office (Project Management). Kirsten believes in Agile approach through & through having led the Salesforce IT org to become agile there. I am working in a team with Kirsten to make working in small teams a reality across PayPal.
  • Allen Olivo. Head of Global Marketing. Allen is a luminary in the industry having led Apple during the Steve Jobs transition back in the fold. He has places like Amazon & Yahoo in his past as well. Like the new fresh PayPal look? That is Allen & his team.
  • Doug Crockford. JavaScript Architect & Industry Luminary! Need I say more?
This is not an exhaustive list. I am sure I am forgetting to mention someone. I could also make a much longer list of VPs, Sr. Directors and Directors that have joined in the last year bringing fresh perspectives into the company. But I should mention that there are many existing leaders (like my former boss who recruited me) that also have this DNA of change.

New Way of Working

I have mentioned before that PayPal is in the midst of a transformation from working in silos in a slow waterfall manner to a lean, startup approach. Earlier in the year I had the privilege of helping to kick off a new way of working on a core part of our business. We took lessons from startups and decided that going forward (this was with David's total blessing as well as mandate) we would not work the way we had in the past.

We gathered our designers, our product folks and our engineers and took over a few conference rooms and began to operate like a startup. Design was done on whiteboards and coded in real time. Usability tests were weekly so the pace was fast and furious. But we were able to try dozens of experiences across desktop, tablet and mobile in the time that would have taken years at PayPal before. Build/Test/Learn became our mantra.

To date we have trained several hundred designers, engineers, product, QA, and other employees in Lean UX methodologies. In addition, we are in a huge transformation project bringing more and more of PayPal into this mode of working every day. Currently I am involved in at least 6 new innovation projects all operating like a startup within the company.

You can read about this approach in my presentation on the Lean UI Stack as well as see my cautions for what can go wrong in the presentation Anti-Patterns for LeanUX.

New Technology Stack 

I am really excited about this part of the transformation. For years PayPal's UI has been built on top of XML/XSLT technologies. If you know me you will no doubt have heard me rant violently about how this technology is the bane of good UI engineering. I won't go into it here, but suffice it to say that the only thing worse than XML/XSLT is a proprietary form of XML/XSLT -- which of course is what PayPal was built on (C++ on the backend).

In the last year the company has moved to Java for all new projects. I am very comfortable with Java. In fact since 1996 I have personally built 4 UI frameworks on top of the Java stack. However, I long ago lost my excitement for using Java in any way, shape or form for the experience layer. Typically Java Server Pages (JSP) is the standard Java solution for UIs. And not very surprisingly this is exactly what was in flight when I joined PayPal.

Knowing that Java was an improvement over C++ and JSP was an improvement over XML/XSLT wasn't enough to satisfy my requirement that UI Engineering (UIE) needed to be able to "Bring great design to life quickly". Elsewhere I have written about how the UI layer is actually the experimentation layer. And that you must design for build/test/learn. Which means designing for volatility and throwaway-ability. Iterating in Java and JSPs just didn't cut it based on my experience.

So with the launch of the new "startup" for one of our core businesses we began to experiment with nodejs. Node is an extremely powerful development environment for applications. Many node modules exist to bootstrap an application. And coupled with the right stack on the UI layer we saw it as a winning combination.

We wanted to be able to iterate faster than we could with the JSP stack and we wanted to be able to bring new talent in the door and have them checking in code within the first few days of hiring. In our mind this was only possible by using open source software as much as possible. Here are some of the choices we made at the UI layer:
  • Dust JS for our templating. We are partnering with LinkedIn as well as contributing to their fork of the Dust JS code on github. Dust is great because it compiles down to JS and so the same templates can be run in the server as well as in the client (page-oriented vs applications).
  • Backbone JS for eventing/model/structure to our applications. Of course underscore JS comes along for the ride.
  • Twitter Bootstrap. For our components. Very nice.
  • jQuery. Naturally.
  • For mobile: jQMobi. Although we are continually re-evaluating choices here.
  • Less. For CSS pre-processing. 
  • Require JS. For module discipline, dependency management, packaging and minification.

However, we were faced with a dilemma. PayPal had invested significantly in the Java layer. It was easy enough to develop with these UI technologies on top of node. But what about the Java/Spring foundation below?

What we did was a rather elegant hack. We added a new ViewResolver in Spring to handle running Dust templates within the Java server stack. We use RhinoScript as the resolver since it is the JS execution engine for the Java VM world.

On the node side, we continued to use it for all of our mockup/prototyping of the actual product. The UI stack sits nicely on top of this node stack. And at anytime we just push the UI portion of the app over to the production Java stack and it runs the same as it did in node. In effect we have made our UI prototyping code be the same as our UI production code.

We continue to investigate the feasibility of using Node in production and working with our PayPal infrastructure team are continuously adding modules to Node to have it operate as a first class citizen in the PayPal environment. In addition, the team has built a node bootstrap that allows you to gen up applications, views, controllers for our node apps -- a lot like the Rails scaffolding.

Another great change internally has been the addition of GitHub to our full enterprise. Nothing like having GitHub as your blessed code repository. All of our node code lives there so the whole org is can make it better.

What has been the result?

  • New UI Engineer hired and within 4 hours was checking in code! PayPal record as far as I know.
  • "Coding is fun again" -- quote from UIE after working in our new  tech stack.
  • Spun up half-dozen new projects behaving like startups using the new technology stack with very little training (if you can google it you don't need a week-long class to teach it).
  • Innovation is happening. We are seeing high levels of collaboration and continuous build/test/learn cycles happening in multiple teams.

New Talent

Last but not least is the talent. I was pleasantly pleased to find good to great talent within PayPal when I joined. Two top engineers at Netflix joined my team and are leading out on a lot of the innovation. The existing members are actually folks I sought out to work for me before and are doing excellent work. I have hired great talent from Yahoo! as well as other places and am looking to deepen the talent further. But I don't want to stop there. I am looking for talent density within my organization and across PayPal.

What about you? If you are a top-flight UIE (frontend engineer, web developer, etc) and it jazzes you to imagine the change we can bring with the resources of PayPal focused correctly, then why not drop me a line. I always have room for top talent. Come help be part of the new PayPal mafia :-)

Contact me now at bill.scott _AT_


Sunday, July 29, 2012

Parenthesis of Forgetfulness

I am reading the excellent book Sleights of Mind: What the Neuroscience of Magic Reveals about our Everyday Deceptions.

In discussing how magicians manipulate attention (looking at the way Apollo Robbins the Gentleman Thief picks pockets) they described a technique called time misdirection. This is when your attention is misdirected in the time dimension. Magicians may introduce delays between the method behind a trick and the effect itself, which makes it difficult to link the two (see pg. 69).

Here's an example from the book:
Imagine that a magician fakes a coin transfer from his left to his right hand, and then opens his right hand to reveal that it is empty. Because there is no separation between the sleight (the fake transfer) and the magical effect (the vanished coin), you may easily conclude that the coin was never actually transferred but remained concealed in the magician's left hand. A more accomplished magician will introduce a separation -- a parenthesis of forgetfulness -- between the method and the effect. For example, after the fake coin transfer, and before revealing his empty right hand, he may reach into his pocket for the overt purpose of retrieving a magic wand, but in fact he is also dropping the palmed coin inside his pocket. Then, touching the magic wand in his left hand to his right hand, he shows that the coin has disappeared.
You can also see the master of sleight, Slydini, using time misdirection over & over in this routine on the Dick Cavett show:

Parenthesis of Forgetfulness

A primary skill of magic is to learn how to hide information. In this case, magicians use the delay of time to take advantage of the way our brains work and misdirect you from the method to the effect. However, when we are designing interfaces this parenthesis of forgetfulness can work against us (often the things that magicians employ, designers have to avoid).

When designing an experience we need to keep the result as close to the action as possible. In other words we have to shorten the parenthesis of forgetfulness.

Performance delays can create this parenthesis. Let's say you want to withdraw or send funds on your bank's web site. If you withdraw funds but it takes a while to show your balance updated in context on the page then your brain will have to work harder to causally tie the action and the result together. One solution is to keep refreshing the user's memory of the relationship by such methods as progress indicators while the user is waiting for the result.

Taking the user out of context through a page refresh to show the result of an action will also create this parenthesis. The time it takes to get to the result (the new page) as well as the change blindness that results from the refresh zaps the brain and takes it to the zone of forgetfulness. A solution to this is to draw the user's attention to what just happened. For example, highlighting the change will work as it refreshes the user's memory and ties the action and result together. But the best solution is to not have action & result span a page refresh. This will remove the parenthesis altogether. Elsewhere I have written about this disconnect before, especially as it relates to change blindness.

What examples of the parenthesis of forgetfulness have you seen in web & mobile experiences? What solutions have you found for these situations?

Saturday, July 14, 2012

My Lean UI Tech Presentation from Open Web Camp 2012

Thanks to John Foliot for putting on an excellent Open Web Camp. It was especially great as we got to host here at PayPal at our Town Hall facility.

Lots of great talks. Check out the presentations at (more to come).

Here is my talk from this morning.

Wednesday, June 27, 2012

Anti-Patterns for Lean UX

We are going full bore on LeanUX at PayPal. This presentation just captures a lot of cautions for our teams. These anti-patterns call out bad behaviors or situations that can become bad which will stifle collaboration.
Lean UX Anti-Patterns
View more presentations from Bill Scott

What anti-patterns have you seen?

Saturday, May 12, 2012

Welcome Crock!

Welcome aboard Doug! Stoked to be working with you again :-)

Interested in where PayPal is headed in the world of Javascript/User Interface Engineering? Ping me. bill.scott at

Sunday, April 22, 2012

The Experimentation Layer

Designing for Reusability

Code reuse has been a major concern of mine throughout my career. I graduated with a computer science degree and was taught the value and beauty of loosely coupled, modular code with a clear separation of concerns. In fact some of my most rewarding experiences have been writing user interface frameworks (a few dozen over the last 30 years).

The same goes for designing experiences. Early on I realized that interaction designs could also be reused. Sometime after reading the gang of four book (Design Patterns) I started thinking that patterns could be used to define experiences as well. It was after that I stumbled upon Jennifer Tidwell's excellent work which would later become the book Designing Interfaces. A few years later, I went on to help launch the Yahoo! Design Pattern library and write a book on the subject.

I mention all this to set the context that I am a huge fan of reuse.

Designing for Throwaway-ability

In my prior role at Netflix, I quickly discovered a wrinkle to the reuse mantra. If you know a little about Netflix you will know that experimentation is the bread and butter of refining the experiences there. The big "aha!" moment came (2008) when I counted the number of experiments we fielded vs the number of experiments that became a lasting part of the site. I discovered that

Only about 10% of the UI code written to craft the experience lasted more than a year. 90% of the UI code needed to be thrown away.

Could that really be right? In some cases we were testing 10+ test cells over a couple of months time. Only 1 cell (or sometimes a combination of cells) was selected as a winner. The winning cells often led to a new hypothesis that led to another set of test cells being fielded (perhaps another 10+ cells). And of those a new cell (or control) would end up as the winner. It doesn't take too many 9 out of 10 losers to start seeing that a lot of code needed to be thrown away over the course of a year. (I am over-simplifying this a bit. There is often reuse between test cells as well as the number of cells fielded will vary.)

Now to a typical user the experience doesn't seem to change much in the short term. User A might be in a control cell (the standard experience). And in the next set of tests User A might still get allocated to the standard experience again. So to User A the site seems pretty stable. However if you plot the arc of the experience year over year you will see some substantial changes occur (think: Amazon, Netflix, Twitter, Facebook) as newly minted experiences are released to the full user population.

The epiphany for me was that for the most volatile part of the experience...

We needed to design for throwaway-ability and not as much for reusability.

What does that even mean? Design for throwaway-ability? It means not over-investing in the scaffolding for reuse. It means designing a way to select code for experiments in a simple grokkable manner (like using the file system or packaging mechanisms for bundling experiments together). It means externalizing the logic for selecting experiments as much as possible. It might even mean copy and paste reuse to make it easy to modularize the experiments. It means don't start with writing components, instead start with creating the experience.


When I had this epiphany it certainly felt heretical. Over time I have come to see the wisdom of knowing when to reuse and when not to reuse. However, lest I be next for a good stake burning for giving others a license to writing crappy code (since one could argue that it will be thrown away anyway ;-) let me provide some nuance.

Reuse is Still a Good Thing

Not all of the UI layer changes with experimentation. The nuance is the UI layer contains different levels of volatility. The code to make a button or render a template doesn't change with an experiment. However, the placement of the button or the contents of the template will often vary from test to test.

For the more stable areas of the experience, the great news is we have a rich set of open source libraries and frameworks that can provide a lot of this reuse right out of the box. Examples include component libraries like YUI or jQuery; templating solutions like mustache or dust; MVC patterns like those expressed in backbone.js; common javascript extensions like those found in underscore.js and so on. In addition each organization will find certain aspects of their user experience that benefit heavily from thoughtful reuse.

The truth is volatility increases as you get closer to the user experience and reuse opportunities increase as you move down the software stack.

This funnel illustrates the relationship of volatility and reuse 
as you move up and down the software stack.

UI Layer = Experimentation Layer

What I want to impress upon you is a simple thought.

The UI layer should be thought of as the experimentation layer

Think of the topmost layer of the user interface code (especially the HTML, CSS and Javascript for experiments) as the experimentation layer. As a corollary

You should always create an experience before thinking about creating a component 

In other words seek to use before reuse. I have found in my conversations at PayPal, talking about the UI layer as an experimentation layer is a great way to shift the conversation to the experience rather than to spending months building components and interface scaffolding. The goal is to get code in front of the user as fast as possible. Instead of starting with reuse, we start with the experience.

What about Consistency?

Now let's be clear, experimentation is not the silver bullet that will create the perfect experience. A lot of understanding about the user, scenarios, and just good solid design principles are needed to make a good start on the experience. Experimentation can often lead to local optima instead of finding the overall experience. But given a good experience, experimentation can allow you to refine the experience in the right direction.

One of the valid complaints that designers will often raise about experimentation is that it can lead to different experiences across the product families as the experimentation often is chasing specific business metrics in a specific area of the business. This can lead to some inconsistency when we allow the experience to be experimented on (at Netflix we call this the Frankenstein phase). There is a natural tension between innovation and standardization and the truth is healthy products will swing back and forth between loosening the constraints and innovating and then pulling back in the reigns and standardizing.

Design standardization (read: patterns, guidelines) can be used to bring consistency. But I have often seen design standards teams position themselves at odds to innovation. And on the converse I have seen design innovation teams totally ignore basic patterns that could be employed. Both are wrong. Consistency without informed learnings during phases of innovation is the dead letter. And innovation without the learnings of the past & best practices is just wild speculation.

Engineering teams will struggle with when to invest in the reuse scaffolding and when to simply create an experience that is quick to build and quick to tear down.

The Bottom Line

The key thing to remember is everything must be in service to the user's experience. It's way to easy to forget this and get in an endless cycles of finessing the UI - which I must remind you might not even be the right experience.

Saturday, January 28, 2012

My PayPal Update: PayPal Rocks.

The last three months have been a blur. The last blog I posted was right before joining PayPal. So I thought it would be good to give an update on my honest impressions about PayPal.

Here are my take-aways.

#1 PayPal is positioned like no other company to change the way people use money. 
I am really excited to be part of changing something that is so fundamental to the way the world works. I sensed that PayPal was at an inflection point. And now seeing the products we are working on has convinced me we can do this and we welcome the competition. At Netflix we changed the way people view movies. At PayPal we change the way people use money :-) I am totally stoked about our business future.

#2 The wisdom is in the crowds! 
I interviewed over 100 different people in the organization in my first 6-8 weeks. My theory was the "smarts" was already in the organization. My interviews confirmed this. In many ways my job is collecting the intelligence, distilling it, and connecting the right people and information together and ensuring that we remove friction, indecision and instead multiply the ability of our makers to make.

#3. Innovation is blooming all around the organization
I think this surprises colleagues of mine when I tell them that innovation is alive & well at PayPal. At PayPal's quarterly Leader's day they have a great tradition of every new leader (director & above) standing up and telling about where they came from, what's unique about them and what they will be doing. At the November meeting it took us over an hour to get through all the new people. I would estimate that 15-25% of the leaders were new. Either through acquisition or new hire. All this new blood brings in change with it. Couple the sparks from the new blood with the kindling of the wisdom in the crowds (to mix a few metaphors) and you have a roaring fire. I can assure you the PayPal you see on the site today and on merchant sites around the world will not be the same paypal you see in the future.

And the competition from our friends at Google, Square and many others is awesome. Thank God for good competitors. This is energizing and refines a company and makes it lean.

In fact I am in the middle of hiring for my head of UIE Innovation. This is a kick-butt role that partners directly with UED & Product in a lean & mean & collaborative manner to burn through as many ideas as possible and marry technology & innovation.

#4 The Technology Platform is being Re-Invented
Prior technology choices on the front-end have hindered productivity. Using XML/XSLT is a HORRIBLE choice. It is a curse that thankfully is being rooted out. We are migrating all of the products to a simpler architecture that supports true client side development. I am happy to report that my team and the infrastructure team are building strong alliances and we have a shared vision that is about making UI development as nimble as possible. We are building towards a rapid experimentation platform.

#5 My Leadership Team Rocks
Ok, so you might say "well you would say this because some of your leaders will read your blog." Regardless, it is true. I have never been as challenged by my leaders to be bold, make quick decisions, lead, allow no excuses to paralyze us as I have been here at PayPal.

I can't speak for the past at PayPal. And we certainly are not where we need to be yet. But I can assure you that we are unified on empowering an army of change agents. And we have the right vision, the right story that defines where we will be.

We are Hiring
How about it? Want to join me in this grand adventure? I am on the lookout for talented user interface engineers. I need strong developers. Strong computer science skills. JavaScript rock stars. HTML5/CSS3 aficionados. If you have the chops contact me at billwscott over on gmail.