4 ms·

There once was a new director giving a presentation in an 'all hands' quarterly meeting about how bad our enterprise stability was. He was absolutely right. I

There once was a new director giving a presentation in an 'all hands' quarterly meeting about how bad our enterprise stability was. He was absolutely right. If we had 3 of 5 weekdays where all major systems ran acceptably, it was a good week. End users hated IT. You could see half the room nodding in agreement. Finally, us minions would be permitted to fix some things. Wrong. He was fired and gone the week after. Real advances came from a director a few years later, who started with a 'we're doing good but can do even better' all hands. His slides showed nice green monitoring dashboards with a few yellow lights, even if '3 working days in a workweek' was still very much a fact. Change was very very slow and gradual, but at least it happened.

ludicity 4y ago
Very similar to the lesson I had taken away when writing the article. The first director you mentioned was essentially making the same mistake myself and many others make - we think that pointing out the problems honestly will allow us to justify spending resources on fixing them. The seemingly correct way to handle this is to lie about the current state of things and push for a change anyway. This is counterintuitive to the sane, because things being good means that there's less pressure to change them. The secret sauce evidently lies in agreeing everything is good so that all those involved can agree to your follow-up proposal to fix the obviously bad system without looking like failures themselves. So the next question is, if change is sufficiently slow (and assuming that you care about whether your work has an impact somewhere), is it worth sticking around? You have to lie a fair bit, which is distasteful, and the time it takes to replace one terrible database could take up 2-3 years of your life. But someone presumably has to do it, and if all the competent people leave, then it'll never get better. Still figuring this out myself.

ludicity 4y ago

Very similar to the lesson I had taken away when writing the article. The first director you mentioned was essentially making the same mistake myself and many others make - we think that pointing out the problems honestly will allow us to justify spending resources on fixing them. The seemingly correct way to handle this is to lie about the current state of things and push for a change anyway. This is counterintuitive to the sane, because things being good means that there's less pressure to change them. The secret sauce evidently lies in agreeing everything is good so that all those involved can agree to your follow-up proposal to fix the obviously bad system without looking like failures themselves. So the next question is, if change is sufficiently slow (and assuming that you care about whether your work has an impact somewhere), is it worth sticking around? You have to lie a fair bit, which is distasteful, and the time it takes to replace one terrible database could take up 2-3 years of your life. But someone presumably has to do it, and if all the competent people leave, then it'll never get better. Still figuring this out myself.