RE: [scrumdevelopment] Re: How many stories can/should be in the backlog?
- Your suggestion (John Goodson) is perfectly sound from a lean development point of view, but it does not address the usually vexing question of the demand for up-front estimates of the project.
No matter how much you, and I, and others agree that estimates are 'best guesses' at best, and are always wrong as soon as they are made, the fact is clients and product owners usually want and need an up-front estimate to decide if the project should even proceed. So how does this lean idea assist in this?
This demand for estimates, which usually turn into quotes in most clients' minds, are then seen as the fundamental success criteria of the project ... Within Time, Within Budget, Within Scope. This always strikes me as creating a built-in failure scenario.
Given the topic, I would like to change it a bit to ask about Feasability Studies as a precursor to project activity. Does anyone have any comments on this topic?
Date: Fri, 14 Nov 2008 22:39:05 -0500
Subject: Re: [scrumdevelopment] Re: How many stories can/should be in the backlog?On Thu, Nov 13, 2008 at 8:30 AM, Luciano Felix <lucianofelix@ gmail.com> wrote:Hi
Well, if are talking about the Product Backlog, you shouldn´t worry about that. To me the product backlog is a list of ideas, with some rules of course (estimated, prioritized, more detailed in the top, etc.), but in the end is still a list of ideas and it doesn´t make sense to put a limit in people´s creativity. If a particular story does not fit the product vision the PO should put it in the bottom of the backlog or remove it, but this should be circumstantial, not some kind of upfront limitation.
If you are talking about the Selected Product Backlog, the list of story to be build in the upcoming sprint, this is something for the team to decide, based on capacity they think they have for the next sprint. Some historical data from previous sprints may help the decision, but again the final word comes from the team.
Scrap the product backlog. It encourages inventory and delay. Build a Minimal Marketable Feature (MMF) set using a product roadmap. It's way more lean and effective than brainstorming cards.
John Goodsen RADSoft / Better Software Faster
jgoodsen@radsoft. com Lean/Agile/XP/ Scrum Coaching and Training
http://www.radsoft. com Ruby on Rails and Java Solutions
Find your ideal job with SEEK Time for change?