Published Monday, 13 July, 2015
Following-up from the last episode, Jelly tries to recover from his failed attempts to explain why he believes that opening up Radar isn’t the solution to all of Apple’s issues with bug reporting.
This causes Jake to note that part of the problem with Radar is that Apple, as an organisation, seems to care a lot less than Google about reported bugs in general. That’s not to say that the engineers themselves don’t care, which becomes obvious when you do hear from them directly, whether in labs at WWDC or through avenues outside Radar, such as Twitter or Apple’s developer forum.
Jelly’s about to release an new version of GIFwrapped, so he’s been thinking about things like deadlines and code freeze and how developers handle those things. So he asks Ben and Jake a few questions about how they approach an app release.
Jake feels like the best approach is to develop iteratively, and aim for regular, short release cycles (like Facebook’s two week cycle). Jelly takes issue with that, explaining that as a single developer, or even a small team, it’s very difficult to aim for that kind of model.
Ben then brings up automation, explaining that part of moving to a regular release cycle requires a change in thinking. Jake also explains that perhaps larger features need to be broken down into smaller pieces, making it easier to stick to regular release schedules.