"Dont' keep a backlog at all."---Ron Jeffries
- I saw this interesting article recently on Ron Jeffries on his XPMagazine
site. In it he suggests "don't keep a backlog at all."
I'm interested in what others think.
I think it's a horrible idea.
The cost of creating the backlog is not something you want to go through
every sprint/iteration. The cost of wrongly doing an item on the backlog
that a user would not have "petitioned" for in a later sprint (but that you
did thinking he would) is likely to be less than the cost of recreating the
backlog each sprint.
At a minimum, it's interesting reading and I'd like to hear what other
Here's the link to the article:
- --- In scrumdevelopment@y..., "Mike Cohn" <mike@m...> wrote:
> I saw this interesting article recently on Ron Jeffries on hisXPMagazine
> site. In it he suggests "don't keep a backlog at all."Hi Mike,
> I'm interested in what others think.
I don't think Ron and I have the same vision of a backlog.
I used to program the control systems for new machines to made tape
at 3M. Pretty much standard practice was to develop a punch list at
the end of a major installation. It was basically an agreement that
once these items were punched off, the job would be done. The punch
list was prioritized by the customer, and worked on in more less
priority order by the installation team, until it was done. That is
how I see a backlog.
The important thing about a punch list is that the customer is
comfortable that all of their concerns about the installation will
eventually get addressed, while the team has a tool to help the
customer to set priorities and a good way to know when they will
really be done.
This is how I view the backlog. Ron's metaphor of a king with a
bunch of subjects, the king making priority decisions on the
subject's requests, doesn't reflect (for me) the concept of a
punch list or the purpose it serves. A punch list is a wonderful
device, widely used and understood. I'll bet if you ever had a
house built or remodeled, you tied up the ends of the contract with
a punch list.