Slashdot Mirror
Fire Your IT Boss
theodp writes "Instead of laying off techies who directly help users, Robert X. Cringely argues that the best place to cut IT organizations is at the top. One of the great problems in IT management, Cringely says, is that the big bosses typically haven't a clue what is happening, what needs to happen, and what it all should cost. He issues the following challenge: 'If you are managing an IT shop and can't write the code to render "hello world" in C, HTML, PHP, and pull "hello world" from a MySQL database using a perl script, then you are in the wrong job.' Even with help from Google, Cringely believes many technical managers would fail this test and should get the boot as a result — you can't manage what you don't understand."
In my particular case not only can he, but he got to be my boss by having my job and delegating responsibilities to grow the business. The same with his boss. Either of them has and can fill my role. That's how it should be and I'm preparing my unders to seize the opportunity. He expects me to do the same so both of us can move up to more responsible positions in a larger business by growing our business. If I fall out and I'm too lame to have juniors step right in, he's prepared to do that until he can bring others up to speed. If the day comes that I must take up his chores I'm ready to give it my best and to be replaced by someone better.
Again, it's about the customer. The customer does not need to worry about personnel details. When I recently I decided to swap out a key person in a customer facing role, I just did. The customer did not ask why because it's implied that I did so in their best interest and with due consideration for the inconvenience. They trust me in this role and they would trust my replacement also. We do what the customer needs done with minimum fuss or the customer hires somebody else. Period.
Help stamp out iliturcy.
If a manager wants to manage someone who's writing C/HTML/PHP code he's going to be much better at it if he knows what C/HTML/PHP code is.
Cringley didn't suggest the manager be the hottest programmer in the organization, he suggested the manager be familiar with the absolute basics.
I often think the better way to consider a manager is as an assistant to those who do the actual work, taking care of the peripheral details of a project allowing the important people to do the actual work.
Excellent point. My manager is an old COBOL & IMS programmer who can't write a word of SQL.
But... he knows what the difference is between OLTP and a Data Warehouse, knows that IO is slower than core, and is bright enough to have learned from us the difference between sorted and hashed indexes, and when to use them, and what happens when indexes get out of whack.
Thus, even though he can't do what his underlings do, he knows what's reasonable and not, and bravely defends us from User Stupidity and Programmer Incompetence.
When he retires, I'm going to cry.
"I don't know, therefore Aliens" Wafflebox1
It's in downtown Chicago. Can't say the name of the company. If you are in the area, send me an email - yes I have 2 positions open right now.
Raise your hand, everybody who has had to dumb down a report for management.
When a manager sees that the average traffic is way lower than the vendor stated capacity and the load is 0.5, it won't do you much good to try to explain phrases like "slashdot effect" or that while database reorgs are running, the boxes run out of juice. It's your fault, and the only way you can convince the manager otherwise is by getting the vendor to admit he sold you underpowered boxes, or the dev IT manager to admit that the internal apps he is responsible for sucks.
So you write an "executive version" of your report, stating how in light of the manager's recent successes, it would augment core strategies to implement a synergistic interdepartmental policy for capacity preparedness scaling, and by the way, that you need $30k worth of IBM/Sun/Microsoft/SGI (strike what doesn't fit) recommended hardware and half a man-month of DBAs assigned to you to do so. If you can give the project a meaningless buzzword name, even better.
A good manager, on the other hand, you can tell straight out that the system is both underpowered and ill-designed for peak loads, and even ask what the chances are of getting funding to fix the problem versus band-aiding it.
Carburetors are something that gets taught to a 15 year old student and is as such certainly within the grasp.
Perhaps you attended a high school with an auto shop class. I didn't and neither did any of my close relatives or friends in my generation. The last carburated car in the US was produced in 1991 which coincidentally was the year I graduated from high school. Every car I have ever owned has been fuel injected. Yes you can find carburators even now but mostly on the smallest and cheapest engines and they are slowly going away. Even motorcycles are headed toward fuel injection these days. The only reason I know anything about carburators is because I'm just kind of a curious geek that way and find them interesting.
Point is, knowing how a carburator works has little to do with the skill set of being a decent manager - automotive industry or otherwise. I've worked as an automotive engineer and I know this from first hand experience. ,
A 'car guy' is not necessarily good at running a business.
Henry Ford and Enzo Ferrari did pretty well at it. So did Ferdinand Porsche, Frederick Royce, Karl Benz, and Walter Bentley.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."