With a flexible estimate, is it true, some say, to choose intervals, for example, from 1/2 to 1.5 days?
As a general rule, it is a good rule of thumb (flexible or not) that your tasks should be broken down no more than 1 to 2 days.
The idea is that if you have more chunks than you have, then you haven't broken the task enough, and you are more likely to skip the assessment and miss it for a longer amount of time than if you broke it. Often times, when you break it down, you find that your initial assessment was off, and since you broke the task down to more specific tasks, your assessment is now more accurate, more traceable, and meaningful.
For tasks that appear on your to-do list shortly, you should pay attention to this, but for long-term planning, where you did not necessarily think about this detail in detail, I think that larger estimates / tasks are not broken down into this function OK.
Here is a link to Joel Spolsky about it. Take a look at # 5 about halfway down the page.
source to share
In my experience, any assessment that lasts more than 2 days is likely to hide serious work that needs to be broken down further. Such assessments have a very high probability of transition. Try to break everything down into smaller pieces so that no single piece is worth more than 1-2 days.
source to share
There are advantages to keeping scores to a minimum. This forces you to break down large tasks into small discrete tasks that can be quickly measured and discussed, which helps move the entire Agile development process forward.
That being said, I almost never follow the "rule" as a hard and fast rule with this kind of thing. I would say that this is a good recommendation.
source to share
My team is made up of junior programmers (university students), and we've found it's generally easier if we break all the big tasks into a bunch of smaller ones. This is associated with more thoughtful thinking, but in the end we are more productive and can more easily measure our progress. It also brings a sense of accomplishment when you have something complete at the end of the day.
source to share
I disagree. If the iteration in a team lasts two weeks, 10 days means 1 day will be spent iterating close (show and say), schedule the iteration and the poker task or scheduling.
When you play poker planning, the command is either geometric or Fibonacci progression for the scores. For example, maps will contain values such as 1, 2, 4, 8, 16 or 1, 2, 3, 5, 8, 13. Each number represents the number of development days for a couple of programmers.
For each card, once the discussion has taken place, each member simultaneously plays a card that reflects their assessment. If the majority of the team agree on the same score, the score is accepted. If there are many variations in the ratings, further discussion takes place (members explain the reason for their ratings) and another round of voting takes place. This happens before consensus is reached.
If a number greater than 8 is chosen, the card is considered too large and the card is reorganized into at least two smaller cards. The reason is that such a large estimate indicates that the map is too large to complete in one iteration, and any estimate is likely to be inaccurate.
Using this method, they take on the delivery responsibilities of the team members that they have made, and for the new team, the estimate becomes so accurate that card transfer soon becomes a low risk.
source to share
You can find a very good post on Agile Assessment and Planning on the agile42 blog: Easy enough, just in time
source to share
There are many good answers here, so I'll play devil's advocate and approach him from a different angle.
There can be a problem with breaking things down into very small estimates (number of hours) when doing things like release scheduling. David Anderson discusses this in his (excellent) book Agile Management for Software Engineering .
Basically, the idea is that for a very small task, the developer evaluates their score with a fair score (say half an hour or doubling) due to a certain amount of ego that would be hurt if the developer failed to complete such a small task in estimated time. These local buffers add up quite a bit and result in a global buffer that is much larger than needed.
This is not a problem if you stick with .5 days at least - basically it was supposed to have some kind of buffer in there, so you won't have to wipe it anymore.
source to share
I feel like there is a bit of information and overlap in this thread ... let me point out :-)
1). The Fibonacci Sequence, which is very useful thanks to Mike Cohn's poker planning method, is an estimate of the "difficulty" of user stories, which Cam said are usually written on cards and entail more than one challenge, at least all of those that will be needed to deliver the story (Ken Schwaber, Alistair Cockburn, Mike Cohn ...)
2) The tasks that are included to complete the Story are usually scored in "Ideal Hours" or "Pomodori" (Francesco Cirillo, "Pomodoro Technique"). If you are rating in Ideal Hours, the rule of thumb is to keep them between 1/2 days (3 ideal hours) and 2 days (12 ideal hours) of work. The reason for this is that the team will have better status information with at least every two days a team member reporting an assignment as completed, which is much more "valuable" than 60%. If you are using Pomodori, they are implicitly "timed" to 25 minutes. every
The reason for keeping small tasks is mainly based on "Empirical Process Management Theory" for which through transparency and regular monitoring and adaptation, you can better check your progress by quantifying it. The purpose of the small tasks is to clearly describe and present in detail what will actually be done without adding too much "guessing" given the natural uncertainty associated with predicting the "future". Moreover, the definition of the result and the shorter time frame allow people to focus on enough “sense of urgency” to make it challenging and motivating.
I'll also catch up to the point of "motivation" and "ego" - from Chris - adding that a good way to get people to commit and motivate is to define the expected outcome of a task, so that you can measure the results after completion and celebrate success. This idea is encapsulated in the Pomomoro technique, but it can also be achieved using the ideal clock for judgment. Another interesting part of the "Pomodoro" technique is that "breaks" are considered "first class citizens" and are regularly scheduled, which is very important, especially in creative and brain activities :-)
What do you think? Best
ANdreaT
source to share