3 ms·

So if I'm reading this right: If a project that uses Agile fails, the participants have no incentive to blame each other. If a project that uses Waterfall fai

So if I'm reading this right: If a project that uses Agile fails, the participants have no incentive to blame each other. If a project that uses Waterfall fails, the participants have an incentive to blame each other. Because Waterfall in the failure case gives people an incentive to blame each other, it's a poorer methodology than Agile. If I haven't confused your argument, then the weakness is that you have not shown that Agile is a methodology in which people have no incentive (or ability) to blame each other if the project fails.

raganwald 15y ago
My claim is that if a project that uses Agile fails, the participants still have an incentive in dysfunctional cultures to blame each other, it's just that the structure of Agile makes it harder to blame individuals. So I am not claiming that A is better than W or W is better than A at shipping software, I am claiming that W is better than A at blaming people for failures. I might have confused the issue by trying to draw a distinction between blaming the process from the inside and from the outside. My claim is that it is much easier to blame a process if you aren't a participant in a failing process. For this reason, it is easier to evangelize a new process of any type as a consultant or as a new face than it is if you are an old hand who may be accused of having been a cause of previous failures. I think this second point is true of any process.
- will_sargent 15y ago
  Can you explain how the structure of Agile makes it harder to blame individuals?
  - raganwald 15y ago
    Perhaps we could pick a methodology and then imagine a project that "fails." Given this scenario, the challenge would be to find a way to blame the participants for the failure. What promises do developers and stakeholders make to each other in the methodology? How do you define "failure" and given that definition, how do you establish a causal relationship between an individual and the result?
    - will_sargent 15y ago
      Well, the great thing about scrum planning is that you know exactly how long it will take and who to hold accountable. http://marcin.floryan.pl/blog/2011/06/nightmare-agile http://marcin.floryan.pl/blog/2011/06/nightmare-agile

raganwald 15y ago

My claim is that if a project that uses Agile fails, the participants still have an incentive in dysfunctional cultures to blame each other, it's just that the structure of Agile makes it harder to blame individuals. So I am not claiming that A is better than W or W is better than A at shipping software, I am claiming that W is better than A at blaming people for failures. I might have confused the issue by trying to draw a distinction between blaming the process from the inside and from the outside. My claim is that it is much easier to blame a process if you aren't a participant in a failing process. For this reason, it is easier to evangelize a new process of any type as a consultant or as a new face than it is if you are an old hand who may be accused of having been a cause of previous failures. I think this second point is true of any process.

will_sargent 15y ago
Can you explain how the structure of Agile makes it harder to blame individuals?
- raganwald 15y ago
  Perhaps we could pick a methodology and then imagine a project that "fails." Given this scenario, the challenge would be to find a way to blame the participants for the failure. What promises do developers and stakeholders make to each other in the methodology? How do you define "failure" and given that definition, how do you establish a causal relationship between an individual and the result?
  - will_sargent 15y ago
    Well, the great thing about scrum planning is that you know exactly how long it will take and who to hold accountable. http://marcin.floryan.pl/blog/2011/06/nightmare-agile http://marcin.floryan.pl/blog/2011/06/nightmare-agile