CWG: Time To Do The Hansei

The Commonwealth Games by all accounts went off well. That it was headed for a disaster will soon be forgotten. If we do get the 2020 Olympics there is every likelihood that we will again prepare for it in the same manner.

A large complex project, involving multiple government agencies, with no option of slippage requires a highly skilled and motivated management team. And since the national image is involved, the team must have access to, and the whole-hearted support of, the highest rungs of the political leadership. Suitable talent should be shortlisted. The UID project, headed by Nandan Nilekani, brought in with the status of a Minister of State in the Union cabinet, should be the model.

Based on the current experience, we need to create a project plan - complete wih check lists (e.g. signs near building housing the ozzie team saying,"Beware of Falling Refrigerators:-). The project plan must be periodically reviewed and updated since financial conditions, technology, security environment, facilty benchmarks, vendors, ,, all will change. The sports ministry should do this, if we are serious about conducting the 2020 Olympics.

Holding enquiries into alleged siphoning of funds, punishing the guilty, assuming any are found, is all very well, but it will not reduce the chances a repetition.

For that we need to do the Hansei.

Lessons from the Commonwealth Games: Deadlines?

The media has been reporting one Commonwealth Games deadline after another being missed.

What should happen when an activity misses its deadline? Should not the activity be declared dead? The outcome of the activity ceases to be relevant after the deadline. Why else call it a deadline?

The fact is these are not even estimated dates of completion. These are just ordered from on high: "Thou shalt complete this task by dddmmyyyy, or thou art dead". It may induce fear. It certainly does not get buy in. People get very adept at passing the buck. Forget about being dead, no one is even held responsible.

Does it happen in software companies? I am sure it does. At one time it used to happen at acmet's Tools BU.

Lessons
Remember, we deal with estimates. Not deadlines.
Estimates must be made by the people who have to do the execution.
Estimate, and re-estimate, on the basis of performance till date.
Do not hide the bad news.
Learn (Do the Hansei). It is better than fixing responsibility.

New Toys

Seth's post reminded me of something that I have seen quite often with software developers.

Software developers want to work on new projects. Given the choice between finishing a project and joining a new project, hardly anyone opts for the former. It is in the begining phases, when the product is still "plastic" , that work seems more like play. It is like being in kindergarten. But if all a person does is move from one begining  to another, the person will forever remain a beginner - no matter how many projects are put on the resume. The person will forever be in kindergarten.

It is only during the later phases, particularly the maintenance phase, that one gets to realize how things should have been done. That is when learning takes place. That is when one starts to become an expert. It takes time and reflection (hansei). There are no daily thrills. You do not get to play with new toys everyday. You have to study your old toy and learn how it should have been made. Then you can go and make better ones.

To Err is Human ...

To Err is Human; to Forgive - Divine

But to Forget is Stupid!

Create, or modify software, and defects will be introduced. Penalize individuals for it and defects will be be swept under the carpet. Team cohesiveness will disappear. So forgive - and forget - the person.

But to forget, what led to the defect being introduced. To forget to learn from it. To forget to put preventive practices in place. That would be stupid!

Do the hansei.