011 How do we do things we would never, ordinarily do? – Lords of Computing Podcast

Summary

Matt and I talk about lessons learned from almost a year of helping transform IT at Allstate. When it comes to scaling up agile and cloud-think the real challenges are in functions other than development, like budgeting, planning, training, hiring, and how the overall IT department is organized. We discuss those topics – esp. budgeting! – and also how to set one’s personal expectations about going on the transformation journey. Then we discuss an upcoming column on mine in The Register on the benefits of small batches thinking.

Listen above, subscribe to the feed (iTunes, RSS Feed), or download the MP3 directly.

Show-notes and Links

  • After a year, the question becomes “can it scale?”
  • How do we do: Budgeting, training, hiring, how do we organize teams
  • We only plan with good information, not bad information.
  • You need to establish an overall vision, but avoid being too specific on tactics. For example, with a claim application, we know the general product, the vertical, the line of business we have roughly an idea of what claims are, who the customer is, and what that experience is like. Delivering a better experience for claims, what that feels like, and how do we measure it – these things we don’t know perfectly up-front, so we have lots of discipline around iterating and experimenting to deliver good product.
  • How budgeting changes in this small batches approach.
  • With a lot of this, you can’t talk someone into doing these things up-front. They have to experience it first hand: you have to walk them through it.
  • “Sometimes ‘nothing’ is a big win.”
  • Coté’s DevOps columns at The Register.
  • Not mentioned, but good thinking to be had in Larman’s Law
  • Matt Curry: @mattjcurry
  • Coté: @cote, cote.io

One thought on “011 How do we do things we would never, ordinarily do? – Lords of Computing Podcast

  1. This discussion is fascinating but not compelling. What I heard you say is that you only plan 2-3 weeks out, you’re figuring things out as you go along, you’re going to do what the customer needs not necessarily what the customer wants, sometimes things aren’t going to get done, and even if you were successful, the rest of the company couldn’t keep up with you.

    I don’t really understand the money discussion either. The big lump of money is to deliver the overall vision (stated above but I don’t think you referenced it in the podcast). Giving out “money for promises” in small increments would lead to larger battles between engineering and the “big evil business”. It would take the responsibility off engineering because they would make shorter term promises that would be monitored by the financiers but would make engineering more unhappy in the long term because they’d have to deliver to promises instead of long term vision.

    When people deliver hardware boards, they have to deliver with the current available “catalog” of part available to them. Later, they can go back and rebuild it with new parts. It’s the same in software, you architect and deliver software so parts can be replaced as new options come along.

    This discussion seems to put the onus on the business to hire a very good, very motivated team because it seems very difficult to determine whether they are going to be successful until they are done. I agree that you need the small experiments to show people that you are making progress, however, the discussions above could be made by someone totally incompetent saying we need more time and how do you determine if you are asking for too much too soon or dealing with an incompetent, unmotivated team.

    My personal feeling is that you do need a certain amount of architecture and infrastructure before you can work in an agile manner – or, at least, what I think of as an agile manner. That’s why open source is so compelling, it gives you a jump start.

    I do believe in this new way of developing software but this podcast was not compelling for me. It sounds like people wanting to have their cake and eat it too.

    I really enjoy your podcasts. Thanks for making them.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s