Scrum is not intended for long running products. It is intended for product mode, where you get an 8 to 12 months or so of build phase–prior planing required–then you switch to a more relaxed way of development.
At $EMPLOYER we have three phases:
Build
User Acceptance testing
Maintenance
All of them are agile but are different enough to be adapted to product phase.
Thing is that you generally don’t know how big or how long lived a project will get in practice. I’ve worked with lots of projects using scrum before, and early decisions to solve short term goals often cut velocity later on as the project started getting more complex.
So, yes short feedback loop is good, but it needs to be combined with larger scale planning as well. Somebody has to think about the bigger picture and see how each sprint fits into it.
Well this does not prove we should avoid scrum. One needs to think and plan before jumping. However, if knowledge about the project is small, then maybe, something agile is required.
I’m simply pointing out that agile isn’t the silver bullet people make it out to be. It carries a different set of trade offs from waterfall, and can often lead to a different kind of mess. Having short feedback loops is obviously good, and I’m not arguing against that anywhere.
What I’m saying is that short term goals have to be balanced against long terms ones and it’s important to keep the bigger picture in mind. Scrum doesn’t account for this problem.
Scrum is not intended for long running products. It is intended for product mode, where you get an 8 to 12 months or so of build phase–prior planing required–then you switch to a more relaxed way of development.
At $EMPLOYER we have three phases:
All of them are agile but are different enough to be adapted to product phase.
Thing is that you generally don’t know how big or how long lived a project will get in practice. I’ve worked with lots of projects using scrum before, and early decisions to solve short term goals often cut velocity later on as the project started getting more complex.
So, yes short feedback loop is good, but it needs to be combined with larger scale planning as well. Somebody has to think about the bigger picture and see how each sprint fits into it.
Well this does not prove we should avoid scrum. One needs to think and plan before jumping. However, if knowledge about the project is small, then maybe, something agile is required.
I’m simply pointing out that agile isn’t the silver bullet people make it out to be. It carries a different set of trade offs from waterfall, and can often lead to a different kind of mess. Having short feedback loops is obviously good, and I’m not arguing against that anywhere.
What I’m saying is that short term goals have to be balanced against long terms ones and it’s important to keep the bigger picture in mind. Scrum doesn’t account for this problem.