Multi-Tasking

Those rumours of my demise on the blogosphere are so very untrue. Yes, I am still alive and kicking. The last few weeks at work have been highly intense. Firstly a new strategic decisions team (SDT) was created which made my life living hell. And then we had a huge lay-off. Surprisingly I survived (at least as of now!). But amidst all this is an interesting story.

START RANT

Is it good to multi-task? Your obvious answer is, of course it is. A less obvious answer is, depends what you are multi-tasking. And probably thats the answer I am looking for.

Apparently my company is in a process of implementing some cost cutting measures. There has been a shift in responsibilities and a change in the way we work. Fortunately (or unfortunately) I am put in-charge of the pilot project which would serve as a model for the other projects to follow. All seems well and good. Finish pilot study in 1 month and revise our model for the other studies (wherever we screw up). Thats my plan and it sounds simple.

Now the SDT comes up with a brilliant plan. Instead of doing the first study in one month, second in the second month and third in the third month, start all together and get all done together by the end of three months. Multi-task, its simple they tell me. By now I am jumping up and down making weird noises like steam coming out of a cooker. When you are in a room with people whose mean age would be 45, the only way you could get any attention is by jumping up and down and making noises like the aforementioned.

No multi-tasking I say, it will kill us. Fortunately I had read this paper somewhere about multi-tasking in the corporate world. I ask for a 10 minute time out, brainstorm with my team and come up with a quick power-point presentation. The core of the matter - "No Multi-Tasking"

Here is what I tell them:
If you have Task A and Task B, both which take 10 minutes to complete.

Scenario 1:
I first complete Task A and then complete Task B. Total Time: 20 mins.
Average time: (Time for task A + Time for task B)/2
= (10+20)/2 = 15 mins.


Scenario 2:
I multi-task. Alternate between task A and Task B every Minute.
Task A completed after 19 mins. task B after 20 mins.
Average time = 19.5 minutes.

So a brilliant guy from the SDT says, "We dont care whether you give it to us 1 minute before schedule or 5 days before schedule." I wish I could have just ignored him, or asked why he was in the SDT. But well, now I bring in my killer dialog.

"How long does it take you guys to task switch?"

And then I go into a sales pitch to save our programmers & testers (indirectly myself). We have a zillion computations going on in our puny brains when we are developing something. Everything from names of variables to data structures and all the other hogwash. When you ask us to jump from one project to another, it takes us time to make that mental shift. I wish they had rules like marriage. You cant get another one without getting a divorce. So assuming it takes 5 minutes to switch tasks, in scenario 1 the average time is 17.5 minutes. For Scenario 2, its 57.5 minutes.

"But if you have a model, all you need to do is implement changes to the model and it will run across all the three projects" says a person of the SDT. By now I actually was jumping up and down making those aforementioned noises. My friend, you can put wheels on your aunt, but does that make her a truck? Have you ever written a program in your life? Have you spoken with your wife and your mother-in-law at the same time? (all these are thoughts going on in my mind). By now my boss gets up and says we should talk about this later (he probably saw my frustration). SDT says, no later, we work simultaneously and just incorporate changes to the model. I say if we mess up the [expletive deleted] model, we mess up three studies simultaneously.

A week later we have a splurge of lay-offs. Someone tells me the SDT exists but the members of the SDT dont. My life gets simpler and I have some time to blog.

END RANT