Monthly Archives: April 2011

Leveling Up Our Front-End Development

Hello World!

My name is Collin. I’ve been at BigDoor from the beginning and have developed many iterations of our Front-End library. I have failed and have been successful. In all cases, the goal is to make something easy, fun and awesome to use. In dev-land this translates to quick iterations, learn from the issues and make it better. That being said, I’m very happy with our recent release of our JavaScript library. There are definitely warts and improvements to be made but I firmly believe we finally have a library to build from.

Engineering Goals
Our Front-End libraries have both short and long term goals. Usually our short term goals are feature driven while our long term goals are for long term adoption. The following long term goals have been driving our development from the beginning.

  1. Must be generic so that a developer can come in and start extending or adding features as they see fit
  2. Support the non-developer community by designing a clean API to support programmatic configuration

In our previous libraries we were more concerned with our short term goals by adding features and iterating over those features to get them right. Those goals still exist from a design perspective but from the engineering perspective we needed to shift how we were developing in order to support these two long term goals.

Past Design
Previously we developed ‘modules’ which provided an interface for a given feature. These modules would provide an API for the feature and handle remote and local communication. We also incorporated jQuery plugins. We used 3rd party plugins and wrote our own to define what feature a node would provide.

If we continued this way, we could never meet our goals. We were only developing the Logic and UI layers. We needed to shift from a scripted feature set to a full feature application with a clear separation of concerns.

Current Design
Before we started a recent custom implementation, we knew we had to provide better separation if we were going to be successful. We spent a few days drawing diagrams, researching best practices, deciding on how to leverage existing libraries and coming up with names for what we were trying to define.

Once we had flushed out our ideas we had to go back and provide some structure for how we could think and talk about it more publicly. We took the multitier architecture model and identified our layers as the following:

  • Communication
  • Data
  • Logic
  • Presentation
  • Configuration

When these were identified we only had a rough sketch of which classes belonged where but we had to start developing due to our impending deadline. However, this was the first time we had a clear separation of concerns and a shared language to work from and couldn’t wait to get started!

We now have the following architecture to work from in order to meet our short term goals as well as quickly and accurately iterate over our feature development:

  • Communication
    • Transport
    • API
    • BigDoor (inherits from API)
  • Data
    • Model
  • Logic
    • Authentication
    • Controller
    • Provider
  • Presentation
  • Configuration
    • Application

Future Design
It would be nice if I could predict the future and tell you what features and functionality we are working towards but I can’t. What I can say is that we are working hard to make this publicly available as soon as possible by reintegrating Facebook Like, Badges, etc., as well as prototyping new applications.

I’m super excited with this release and couldn’t be happier to be a part of this team. I can’t wait to see what we can come up with next!

Goodbye World!

–Collin Watson

Dirty Words

We wanted to share a great post on AdAge Digital from Judy Shapiro, the Chief Brand Strategist at CloudLinux. In “Why ‘Commerce’ Shouldn’t Be a Dirty Word in Tech Marketing” Judy explores social commerce and, how in it’s current state, the biggest piece missing is the sale: “Why does it seem like “cool” marketing technology and “commerce” are mutually exclusive?”

She makes several really great points in her post – we of course had to highlight her mention of BigDoor:
“Some of the cleverest tech companies are recognizing the huge market potential of merging local, mobile and social to drive commerce……take the example of a company called BigDoor. This is a tech company that creates mini toolbars based on gaming theory so every action lets visitors earn points redeemable for products. It’s the first toolbar I have seen that drives commerce forward (double hooray since most mini toolbars just enhance the share function).”

Thanks to Judy for the mention! You can check out more on her blog here.

Sales Force Gamification

One of our pals at SpectrumDNA, Vince Beerman, is a member of the Software Advisory Board at Software Advice and he recently wrote a byline about Sales Force Automation and Gamification, “Sales Force Gamification = (Automation + Motivation + Socialization)^2.” Vince discusses the “coming gamification of Sales Force Automation” and walks readers through intrinsic and extrinsic rewards. He uses these as examples to illustrate how rewards can relate to motivating employees to learn new systems, socialize with peers, and simply get work done. We agree with Vince – gamification is all about finding the right motivators for your audience, promoting desirable behaviors in an enjoyable way even if it’s just something as simple as cleaning up your inbox!

Gamification and News

This past weekend, developer Brian Immel and I traveled to NYC to participate in the Hacks/Hackers “News + Gaming” hackathon and talks. We had a great weekend talking with the journalists and developers who attended, and look forward to more discussions about how gamification can help news organizations increase user engagement and retention.

The Saturday hackathon portion concluded with a few teams earning a variety of prizes we brought to NYC. Julian Burgess won the event with a quiz game called Whose Headline that asks players to match up headlines with the publications from which they came. The prototype sparked discussion of enhancements and potential alternative uses immediately. Other winning ideas included: Headlines2Headlines, a game based on Apples2Apples that has players match headlines to adjectives representing sentiment; Context, an application that allowed users to annotate articles with sources that confirm or contradict statements; and Chutalkingbout, a game based on predicting Twitter trends. (Special thanks to Discovery Bay Games for providing some of the prizes!)

The panel discussion the previous night covered the integration of game experiences with the news. The potential for innovation in this area is great, both from the perspective of creating games that help inform and explain current topics, to that of gamifying news content experiences to help improve engagement and loyalty.

While there was clearly disagreement about the use of gamification to increase site/application audience metrics, I felt the arguments generalized these solutions to the most simplistic forms, not accepting that user recognition and virtual achievement can be valid tools to improve an online experience. Providing a sense of accomplishment or community status and reputation can be a great motivator with real value to the recipient. The suggestion that online loyalty programs and game-like experiences will lead to exploitation of an audience is disingenuous to me – unless you truly believe content consumers are unable to make choices themselves.

Yes, there will be gimmicky implementations that try to cover up the fact the underlying content or service have no real value. And gamification won’t solve the world’s problems. But when done well, the strategies involved can improve the long-term viability of a content site and create much stronger relationships between the brand and its audience. The virtual and real world rewards must be contextually relevant and meaningful, and the publisher must be prepared to iterate and learn from the program and its members. That takes planning and commitment, but as we tried to convey at the weekend event, we’re here to help and look forward to working with you to take on those challenges.

Thanks again to the Hacks/Hackers team and to the CUNY Graduate School of Journalism for letting us participate in this event. Hope to see you all again soon!

– Roy

All Systems Go!

Below is a copy of the letter we emailed to the BigDoor Publisher Network this morning detailing the recent impact from the Amazon Web Services outage late last week. As a company transparency and feedback are crucial to the work we do here and as such we thought it was important to share this letter more broadly on our blog.

As a valued BigDoor publisher we wanted to take this opportunity to thank you for being patient with us through the latest issues stemming from the Amazon Web Services outage. We’re happy to report that as of Monday morning, all customer facing systems are back up and live and have been since Friday at roughly 1:30 p.m. PST – with a few exceptions that caused our API to intermittently go down.

To add a little more clarity as to what happened and how it may have impacted you, on Thursday, April 21st at 1:30 a.m. PST Amazon Web Services EC2 service failed. The BigDoor team immediately jumped in to action but over the course of the next twelve hours our more than 250 live publishers were affected by the AWS outage. We spent a better part of the day in constant contact with our publishers (via Twitter, emails and phone calls) trying to keep our level of transparency about this situation. Our team definitely felt great frustration with the lack of communication coming from Amazon Web Services – so much so that Keith Smith, our CEO, penned a guest post for the Seattle news site Geekwire. We’re not ones to point fingers and lay blame, in fact we often talk about how much we love AWS; we made the decision early on to host everything we do with Amazon Web Services. In the ten months since we launched the public beta of our free, self-serve gamification platform we have handled over one billion API calls and that simply would not have been possible with our small team and limited budget without AWS.

When it became clear that AWS was not communicating with the urgency we felt was necessary, our team quickly stepped into action so BigDoor’s API and publisher accounts were up faster than most systems that were affected by this nationwide outage nationwide outage. While it’s true we were able to fail-over to backup systems to a degree, over the last year there were additional fail-over capabilities we chose not to implement due to time and cost – this is the balance we’re always trying to make as a startup.

As of today, AWS has yet to recover our last critical system, a service that is key to managing JavaScript components. Since that service has yet to recover, dependent services on are not functioning properly. AWS continues to communicate that we’re hours away from complete recovery, and at some point we may need to bite the bullet and set a new course for ourselves with regard to this server instance (at relatively high cost for us). For now, we’re still in a wait state. We’ve decided to keep allowing logins on the site because other critical functions are working as expected (changes to your underlying economies are being saved properly to the backend systems, for example). Rest assured all of your data is backed up in addition to the redundancies that AWS has provided, in multiple locations.

We’re hungry for feedback on this outage – it caused a massive shockwave for all customers hosted in this AWS Availability Zone, and certainly cost BigDoor a great deal of time and money, but it’s hard to gauge its effects across the board for our publishers without your candid feedback. If you haven’t already, please do let us know how severely this outage affected you, anything you think we should have been doing differently, and what you did to mitigate. We try hard to learn as much as we can from these things, and your feedback is essential in that process.

On a positive note, some of you may have heard our news a few weeks ago when we announced the launch of the BigDoor Engagement Economy, a way for sites to engage their users while monetizing their content. The economy is comprised of rewarding user actions with both status and a branded virtual currency as well as an innovative new engagement mechanic called Quests. BigDoor Quests allow a directed engagement experience for a publisher’s users that reward them with additional status and virtual currency. Quests are a new performance based ad format that is being sold on a Cost Per Quest (CPQ) basis. This is one of the BigDoor monetization moments that will make gamification a profit center and not a cost center for you the publisher. While this program is currently in private Beta we encourage publishers to contact us if you are interested in learning more.

Thank you again for your continued patience and support through all this.


Jeff Malek
Co-Founder and CTO