Freigeben über

the bouncing of the bugs

  • Artikel

At the end of a software development cycle, the development and test teams are locked in a clash of the titans. On one end, the test teams are finding bugs and entering them into our database. On the other end, the development teams are tracking down the cause of the bugs and squashing them. For a product like Office:Mac, you could say that the test team has an advantage over the dev team. Dogfooders (who are employees like me who are using the daily builds of the apps for our own real-world use) and beta testers are also reporting bugs. ZBB (zero bug bounce) is that magical day when, even for just a second, every bug in our database is either (a) fixed, (b) brand-spankin'-new, (c) not fixable (we have a dependency on something else -- Apple making a fix, for example), or (d) postponed to a future release. In practical terms, this is the day when the incoming bug discovery rate is lower than the bug fixing rate from devs. It's a major milestone towards being able to actually get your product out the door.

As a user experience researcher, I'm mostly not involved in this process. I submit bugs that I experience when I'm dogfooding, and I help the teams prioritise the incoming bugs for the bugs that have a user experience impact. Mostly, though, I just get to take advantage of their efforts by using the daily builds, which just get better and better every day. Soon, they'll get to start taking advantage of the work that I've been doing for the past few months, as we define the next big thing.

Comments

  • Anonymous
    October 04, 2007
    Nadyne, As a UX-er, do you handle bugs like when you have a multi-master document with say 35 chapters (sub-documents) open, the document map is unreadable (it becomes like greeked text) on Office 2004? Is this fixed in Office 2008? I can provide a sample document for you (the oWASP Guide 3.0) which is 330 pages long, has around 100 level 2 headings, 400 or so level 3 headings spread over 35 sub-documents - i.e. 35 level 1 headings. This number seems to be the problem. thanks, Andrew