What is the optimal minimum time to work on tasks for a single project? [closed]

Question

As I'm getting more organized, and getting more projects, I'm starting to wonder how to best manage time between them. Especially in regard to how I give estimates to the PHBs.

So if the project is broken up into small tasks, say 2-3 hours each, then is it just as productive to work on one task per day as opposed to spending 6 hours a day and working through a good portion of 3 tasks.

I realize that to a certain extent this depends on the project, but I also think that it might be better in general to work on components together so that things stay fresh in your mind.

That said, it means pushing back other projects a week or two in order to devote 6 hours a day towards those tasks.

Just trying to think this through, and thought I'd get some input.

Answer 1

I think the best way to breakup tasks is to split them into small tasks 1-3 hours a piece. This way you can focus on completing small pieces without interruption (hopefully). Ideally you can work on multiple small tasks from the same project all in a row, this saves you the context switch. Unfortunately, we don't all work in an ideal setting (in fact most of us don't).

Managing the tasks is a matter of assigning priority to them. This is how I do it, in descending order of importance.

1. Order of dependency. Obviously some tasks have to be done before others, do those first.
2. What the boss/client wants done first. Once you know which tasks need to be done first, you can figure out which tasks you need to do to satisfy this requirement.
3. Tasks that free up other to complete work on the project. This helps keep things moving, its not always the most optimal for you, but you work in a team with other programmers, testers and clients. If 6 people are waiting for you to complete one task and 3 people are waiting for you to complete another task, do the task for the 6 people.
4. Related Tasks. Group related tasks to avoid context switches and get things done faster.

Summary Always break your tasks down into small units. An hour of work is ideal. Complete your tasks in an order that allows the projects to be completed as quickly as possible. This has the side effect of decreasing the number of active projects you have at once, which is a good thing - though it usually just means your boss gives you more.

Answer 2

There's a significant discussion going on about the effects on your brain when you do a context shift between disparate tasks in Pragmatic Thinking and Learning, something that inevitably takes place when you work on one project and then jump to working on another project.

You should read up the Human multitasking wiki, but from a perspective that your not doing multiple tasks at the same time, rather that your taxing your brain to load up several large contexts in a single work session.

Here's some excerpts taken from the Wiki:

Some believe that multitasking can result in time wasted due to human context switching and apparently causing more errors due to insufficient attention.

When people attempt to complete many tasks at one time, “or [alternate] rapidly between them, errors go way up and it takes far longer—often double the time or more—to get the jobs done than if they were done sequentially,”... This is largely because “the brain is compelled to restart and refocus”.

People have a limited ability to retain information, which worsens when the amount of information increases.

Without much background info, PT&L defined this analogy of your brain's context as fast and accessible storage for your consciousness where you could say that your current context is something like a page file. Dumping a large page file and loading a new, significantly large page file is described as a physical feat that can be performed for a limited number of times per day and that performances degrade every other time. Basically, you end up with a fragmented context, neither are you on the current project, nor the previous one.

Answer 3

There are big picture people (ones who see the interconnections and what the end result will be) and little picture people (ones who need to understand each and every phase until the "big picture" comes together). It is best to take advantage of both of these personalities.The lp will be more focused and detail oriented for the small parts of the program and the BP will have a better idea of how things will need to be tied together so as to not leave out any crucial steps. Like any good organizations, tiers are the best. I would have 2 persons focused on the big picture and end result, 3 coaches managing 3 little picture people who are focused on every detail in their scope. The coaches help the 3 specialists to collaborate there sections to fit the bigger scope and transfer the information for the overall picture to them. The coaches also work with one of the big picture people to bring the 3 groups together. The one remaining BP will be the one who goes down the punch list, ensures everything went as planned, no missed information and provides the finishing touch.