Introducing Mediafly Airship

I’m proud to have worked with my colleagues at Mediafly to beta-release Airship, the web-based content management system that we’ve been working on for almost a year.  It’s been a fantastic experience conceiving of, designing, and working with them to build the system.  I’ve learned a few things about introducing new products during this process.

  • Speed is essential. The engineer in all of us on the team wants to build a robust, scalable system up front. But doing so has gotten us into trouble in the past. So, this time, we opted to build super-fast, and worry about scalability when we need to. After all, while we may have a guess as to what the scalability challenges may be 6 months from now, we could be very wrong. So it’s better to kick the can down the road at this early stage.
  • Saying no to feature requests is hard. Very very hard. Especially to ones that are good ideas, but are not a part of the Minimum Viable Product (MVP).
  • Frequent meetings and checkins are important. We met 2-3 times/week in 5-minute standup meetings, and informally lots of times/week to hash out ideas.
  • What was hard in the old system is easy now.  With our old Publisher system, getting the media into the system took ages. Users had to use a user-unfriendly “filechucker” that we wrote years ago, to incrementally upload files. Or, they would have to paste a URL. This time we used highly developed open source tools, and as a result, importing is as easy as dragging and dropping into your browser.
  • What was easy in the old system is hard (for now). Assign security groups to a large number of files was relatively easy. Now, (at least in MVP), it requires making changes to each file independently. While this is a short term solution, it is incredibly error prone and time consuming.

Lots more thoughts coming soon. In the meantime, congratulations team, and keep on flying!