Does the 80/20 management rule apply for developers?

Question

Jeff's recently published article is related to the First Fit Decreasing algorithm of time management, which talked about the Pareto principle (or 80/20 rule) of time management, i.e. 80% of the work that we do in 20% of our time.

Now we all heard the quote programmer:

The first 90% of code accounts the first 90% of development time. The remaining 10% of code accounts for the remaining 90% of the development time.

But all the jokes aside, often as if 20% of your code should do what you want, and the remaining 80% should handle exceptions ... does the 80/20 rule really apply to developers?

Does anyone have examples of why he / is not applicable to us?

Answer 1

I believe that Hofstadter’s law applies.

It always takes more time than you expect, even if you take into account the law of Hofstadter.

- Douglas Hofstadter

In a more serious note, pay attention to Critical Chain Management . He recommends giving two ratings for each step of your project. One of them is an optimistic assessment that you are almost 50% sure that you can meet if everything is correct. Another is a more realistic assessment, which takes into account lost time and mistakes (my rephrasing, do not blame the author). Over time and a few projects, you will find out which estimate is more accurate and how much. It depends on the developer, so you need to keep track.

Answer 2

absolutely! 80% of my time was spent on stackoverflow.com and 20% actually worked.

oddly enough, my performance is the same as ever ...

... the same as before!

; -)

Answer 3

2 hours of writing unitats and demo functions for your clients will save you 8 hours of debugging and rewriting early.

Answer 4

In my opinion, Kozarchuk is right:

The problem is not so much in poor time estimates as in poor / impossible area estimates.

Displaying the result or result layouts for the client / manager as early as possible when checking the accuracy of the codes leads to a better understanding of the goals / requirements.

Remember: A project is successful if it “makes customers happy” when it is done, and not when it meets the requirements known to the analyst when the project was originally launched.

Naturally, this means that a " moving target " is the rule, not a bad one, and there is nothing to fear. It also means that I, as the project manager / architect, must ensure that the cost of changes in scope can be / will be transferred and covered .

How it's done?

• Demo early, demo often (users and their managers in the same room)
• Change Queries Mentality. (Thus, the client knows what the changes are and what the changes are, and therefore the client can use them to reorient their project a la carte).
• Be honest, talk with the client and the developers .. and make sure that they also talk to each other.

Does it always work? NOT

Answer 5

Why are you even asking about the 80/20 rule? You have correctly specified rule 90/90. You already know that 90/90 rules apply to developers.

(Sorry for the answers, not the jokes.)

Answer 6

I spend 20% of my time on what I want and 80% refactoring.

So yes, if you think that it "works" in the first 20%. But, that the last 80% makes it reusable, it is worth supporting and using the pleasure (and not the burden) in the future.

Answer 7

The Pereto principle applies a lot to developers. Some say that 80% of the work comes from 20% of the developers. In addition, 80% of errors are generated by 20% of developers. In addition, 80% of the functions are used by 20% of users. Here are the examples I've heard of.

Answer 8

I'm with Bill Lizard. It ALWAYS takes longer than expected due to very unexpected things or, perhaps, things that were not taken into account.

Answer 9

Yes, the 80/20 law applies to development, but you have to interpret it differently:

• The first 80% of the code is executed in 20% of cases.
• The remaining 80% of the time is not enough to execute the remaining 20% ​​of the code.