It's been said that today's glamorous militia leader (think Che Guevara) is tomorrow's head of city planning and infrastructure development (think - your Dad).  The same thing can be said of platform developers, as they exit the protyping stage and prepare the "production-ready" environment.  Today's glamorous designer is tomorrow's anonymous error handling management system engineer.  

That's right.  Now that we're rapidly entering our commericalization phase (watch for an exciting announcement on this later today), what's also happening is you're seeing the results of a year-long effort that has embraced not only the "big data" arts of data gathering, curation, analysis and presentation, but equally, the under-loved areas of user experience design: load balancing, flood management, query and index optimization, settings management, and error handling. 

For many businesses, this transition can be extremely painful.  Some developers simply don't want to lay down their arms and remove their beret - they simply weren't made to transition their skills into error handlling and query optimizing.  

Wise managers will find a way to buy these engineers a ticket to Bolivia, so they can continue their work as revolutionaries.  Dumb managers will allow them to push off the necessary back-end grunt work and continue the glamorous work of making the front-end ever-more shiny - a move that I know from bitter experience will lead to an extremely well-polished slippery slope down.  

[Entrepreneurs, please take note of the above advice:  Sell your vision, then build the city, brick by brick, - sewer system and all.  You will create far more lasting value than simply putting up a pretty facade.]  

At Heardable, we took the decision to start focusing on "invisible" stuff like error-handling and optimization about twelve months ago.  Today, we're (quietly) launching a highly modular system that has one of the nicer error handling systems built into it that you're likely to (not) come across.  

I say (not) come across, because we're trying to ensure that when a user requests a piece of data that doesn't exist, or needs to be updated, that we do as much of this automatically as possible.  As with most defensive systems, our goal is to not be seen to be doing what we're doing. 

Has this paid off?  Yes.  Several times in the past week,  we've had too many requests for information.  As a small, fast-growing company, our only solution to this is to expand our capacity, and push our business to the cloud, which we're doing as fast as we can.  But there are going to be times when the perfect storm occurs and a system needs to kick in to take care of the user - i.e. a "fail whale" (in our world, a simple alert page) needs to appear and tell the user what is going on.  

Fortunately for us, this is happening less and less because we started in on our planning early, and moved the UX to the forefront of what we focus on daily.  Right now, we probably spend as much time removing and handling conflicts and delays as we do adding capabilities, which feels like the right balance.  

I like to think we're winning the battle.  If you fail to see our version of the fail whale appear, you can be confident that we're winning as well.  if you do - our apologies - and please know that we're working hard to make those appearances less and less.

Note: For all the "fail whale" talk above, we don't have one.  Our current alert page is just that - a simple orange and gray message that lets you know we're ove the limit.  Could this be improved?  Sure - if you have any great ideas about what Heardable should adopt as its "out of connections" Fail Whale, please email me at sharp@heardable.com.  Hand-drawn cartoons, creatures and animals will certainly be given strong preference for inclusion.  We'll even add a credit if you wish.

*You need to have been a regular on Twitter two years ago to get this reference - a big white cartoon "fail whale" designed by Yiying Lu, held up by birds, that used to appear whenever Twitter's servers were overloaded with "Too Many Tweets". The engineers at Twitter did a stunning job of focusing on managing the user experience and adding and improving that infrastructure - and deserve a large amount of the credit for the company's considerable sucess since.  

Hopefully, we're headed down that same path.  Stay tuned.

• March 2013
• 2012
• 2011
• 2010
• 2009