An interesting point, though. Scrum, and sprint planning, surely must depend on having a reasonably stable, albeit short-term, Product Backlog, that can be held at arms length away from the developers during a sprint. It seems to me that if there is a reasonably continuous stream of new requests, the 'out of sprint' stuff', then a development approach that is much more pull driven is more appropriate; take it as it comes, join the queue, we're working steadily and will do your job when it arrives at the head of the queue (excepting the situation where it does have a higher priority than existing queue items).
 
Of course, the question of how efficient and effective the developers are does arise. And how can we measure that efficiency and effectiveness. And how can we evaluate it, for the purpose of improvement.
 
Maybe these are questions that can be addressed to the Kanban-ers :) ... they are serious questions.

 
Regards,
Roy Morien
 
To: scrumdevelopment <at> yahoogroups.com

From: alandd <at> dayleyagile.com
Date: Tue, 31 Aug 2010 21:17:41 -0700

Subject: Re: [scrumdevelopment] Re: What to do when the team is not firing on all cylinders?

 

Good question.  Scrum might be appropriate, depending on the reasons why most of the work is "out of sprint."  And if that "out-of-sprint-ness" is something that hurts business purposes.

Alan

On Tue, Aug 31, 2010 at 9:14 PM, Roy Morien <roymorien <at> hotmail.com> wrote:

 

Going back to basics; if a substantial majority of the team's work is 'out of sprint' work, then is Scrum an appropriate development framework to be following?
 
Regards,
Roy Morien
 

To: scrumdevelopment <at> yahoogroups.com
From: bachans <at> gmail.com
Date: Tue, 31 Aug 2010 09:40:55 -0700
Subject: Re: [scrumdevelopment] Re: What to do when the team is not firing on all cylinders?

 

Michael ,
Great post !!

On Tue, Aug 31, 2010 at 8:14 AM, Michael <mikeabugow <at> yahoo.com> wrote:

 

Alan,

Agreed, this is a case of education and fear. I have discussed this with the team in multiple retrospectives, and their group opinion is they are doing fine because they are getting so much work accomplished. Even though, most of the work is out of sprint,

Is your objective to stop out of Sprint work or to have a more effective way of handling the new stories that comes up. How long are your sprints, intention of the question is to explore whether the Sprint duration has anything to do with out of Sprint .

which they set aside story points for every sprint. (My only victory thus far as their SM.)

We started tracking out of sprint work so not only the team, but also so management could see how little actual planned for work is being accomplished during the sprint. This tracking, unfortunately become the accepted norm.

Mike

--- In scrumdevelopment <at> yahoogroups.com, Alan Dayley <alandd <at> ...> wrote:
>
> This appears to be a case of the team not knowing what they are missing. In
> other words, they don't see a problem to solve so they refuse the solution
> as not needed. Education is the answer and may take a while.
>
> Another likely issue is that they fear application of more control around
> the interruptive work. If they choose to track the flow of the
> interruptions that means they sometimes will have to stop accepting another
> task. Maybe they fear having to work with and regulate incoming requests.
> Again, education of the team and the people making requests is necessary.
>
> I find two major sources of objections to pair-programming.
>
> First, many individuals like to be specialists and see value in being The
> One with the answers for specific needs. They see specialization as job
> security. This fear has some basis in fact and has to be addressed by
> providing confidence that their individual value is secure as possible, even
> if someone else also has similar skills and knowledge.
>
> Second pair-programming is seen as inefficient. Two people working on one
> task is percieved as wasteful since a second task by the second person is
> not being worked on. Education about pair programming benefits in code
> quality, cross-training, culture and team building is needed there.
>
> A quick approach to get past the education need is to couch the new idea as
> an experiment. For example, if they still object to pair-programming,
> suggest that the team try it for just one sprint. One experiment is worth a
> thousand or more white papers and a million presentation slides!
>
> Alan
>

> On Mon, Aug 30, 2010 at 10:53 AM, Michael <mikeabugow <at> ...> wrote:
>
> >
> >
> > Hi All. I have relatively small Scrum team that is responsible for database
> > design, and maintenance. Unfortunately, this team is treated as an
> > operations type team, but is still expected to run as an Agile Scrum team.
> > Greater than 60% of their work is out of sprint tasks. I have suggested that
> > they try incorporating Kanban as part of their development cycles.
> > Unfortunately, neither the team, nor their P.O. sees this as a viable
> > alternative to having constant out of sprint work given to them.
> >
> > Also, each member of the team is considered a specialist in their
> > day-to-day work by the other team members. I have suggested the team try
> > pair-programming so that they each can pick up any task at any time. Again,
> > the team has not seen the value in this.
> >
> > Any suggestions on how I can help this highly specialized, and constantly
> > distracted team?
> >
> > Thanks,
> > Mike
> >
> >
> >
>

__._,_.___