Move Fast and Break Things

When back in 2006 I started to work on UIQ 3.0 and Symbian OS 9.x, I was impressed with the architecture and design of Symbian. Coming from a fast-moving mobile app-development company to work on Operating System that was powering 80% of world’s smartphones at that time was an exciting challenge. What struck me first day that everything that UIQ and Symbian did they did in the right way. Everything you read in books and blogs, articles and white papers was applied at Symbian: Symbian OS was object-oriented from the ground up and had vast sophisticated hierarchy of classes with awesome documentation, development tools were mature and powerful, every aspect of software development process was also done in a right way – we used Agile/SCRUM, great source control system (Perforce) and bug tracking  software (Serena). Every feature was properly designed, reviewed, implemented and tested. When we fixed bugs – we looked at the spec and only then talked to PM/Architect. And yes, specs were actually up-to date (mostly). Every code check-in was core reviewed, had accompanying Unit Tests and automated testing suite was executed (STEAM). However, in early 2007 iPhone came out and it became clear that Symbian will soon become obsolete unless major changes will be done.

Symbian

Copyright Alex Kolesnikov

So why did Symbian and UIQ failed during transition to capacitive multi-touch screens? As with any major disaster (think collapse of USSR) there are many reasons that lead to a quick demise, but I think the most obvious and biggest reason of failure in case of Symbian was that it could not embrace change fast enough. Going even further, I believe that the fact that Symbian engineers, architects and management were used to near-perfect release cycles and great engineering systems, they were all reluctant to embrace the change. Why would you change a great engineering infrastructure? Great engineering culture that was working so far? Being used to a predictable and solid engineering practices inhibited change.

Move fast and break things. Adopt or die. Adopt chaotic, hectic world of Web 2.0 and cloud, when you ship or rather launch every week. Symbian, good bye!

 

7 thoughts on “Move Fast and Break Things

  1. Tomek

    Well, that and the strong hold that the mobile operators had on the market. Apple worked with AT&T, that were much more liberal than say vodafone.

    Reply
    1. andrei Post author

      That too, Tomek. Another fact that is quite interesting is how many mobile OS were released in the past 8-10 years and had popularity: Palm OS, Symbian OS, Windows Mobile, iOS, Android, Windows Phone, Blackberry OS.
      And it seems like every OS was popular for 5-7 years but then died.

      Reply
      1. Sergey

        Partly agreed because Symbian is just OS. For example, UIQ 3.0 in compare to the Nokia’s S60 was quite young platform. And as for me, it was much mo easy to work with UIQ 3.0 rather than with S60.

        Reply
  2. Dmitry

    “it could not embrace change fast enough” – something like that.
    There was a good article (in Russian) about Nokia destiny recently (http://habrahabr.ru/post/171325/). So it said “bad management”, one of visible outcome of which was that “it could not embrace change fast enough”.

    Reply
  3. Sergey

    Hi Andei =)

    On my opinion, Symbian died at first because it was bought by Nokia – this is why died UIQ first and then only Symbian – a lot of different things happened.

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *