Tom Pittman's WebLog

2022 September 23 -- Audit Trail

The Good Lord gave me an above-average ability to work through the implications of cause and effect, and a desire to understand How Things Work. I rather doubt anybody can be an effective computer programmer without both, but the point of today's post is the desirability of eschewing certainty -- at least in public. My friend Phil often would say of himself, "Often wrong, but never in doubt." For myself I amended that to "Occasionally wrong,..." because I tried to learn from my mistakes (and even more from Other People's Mistakes, see "Meditations on the passing of my father" some 19 years ago), so presumably I make fewer of them as I grow older. I never did any count, so it's only a presumption.

As much as I like (my modification of) Phil's aphorism, it only works in private. I really hate keeping two sets of books, my memory isn't good enough to remember which is which, so I do not propose to begin now. The problem is that people -- myself probably included -- really hate being told what to believe, and everybody makes up their own mind for their own reasons, not all of them as logical as I imagine myself to be. Even if they are logical, but are working from different data, they can and will get different results. I was not allowing for that. GR strikes again.

One problem is that I watched two dozen students last year who did not want to be there, but some of them (IMO falsely) said they did. I had a lot of trouble because I am not used to working with people who lie to me. I take what people say at face value, and that didn't work. But I did see what kinds of behavior they engaged in before they finally checked out. I worked out a causal scenario where that could happen (assuming that truth is not very high on their agenda, which seems to be the case in this country). It goes like this:

1. Computer programming is hard work (which it is)

2. Some people have fun doing it, so they are willing to do the work (like me and a few of our students)

3. Most people do not like it and do not want to be here (which the data supports, despite prejudice to the contrary)

4. The Director offered them an opt-out alternative, but it was still too much like programming

5. On Zoom the teaching staff only sees what the student is doing when they are logged in and showing their screen; it is not obvious to them that we can also see the programs that the computer runs (and we also know who ran them)

6. It is a known fact (my blog some 9 years ago quoted the idea in a Koontz novel, which tells us he believed it was typical) that students think they can bamboozle the teachers (but actually cannot in math and CS)

7. Therefore, a clever student might reasonably assume that turning off screen-share (or not even logging on) and claiming to be programming is less net effort than the non-coding alternative in the same class

Every one of the students who did this last year eventually either transferred out entirely, or else elected the alternative (and did minimal work there). Therefore I reasonably inferred that shutting down the Zoom communication is a good indicator of disengagement. The trouble is, the Zoom disengagement is private data, seen by me and mostly nobody else. Since we all live in Athens (where Diogenes carried about a lantern seeking but not finding "an honest man") I might reasonably suppose that the student claim is a lie, but the admins who hear me say it, they might could make the same claim of me (or at least think it). I need to think of this kind of interaction as in a courtroom trial: if there isn't enough evidence to convict, then don't make the claim (out loud). This happened to me this week: I saw the screen blacked out, and on previous occasions when the screen was black he also did not respond to voice or chat, so I didn't bother to try again, and this time the on-premise staff observed there was nothing in the chat log (no audit trail) and made the implication that my claim of disengagement was false.

When I proposed this "Teach Java" thing, I made it clear I can only do this for students who want to learn. That's still true.

Well OK, Phil's "never in doubt" line needs to be less public as a general rule, sort of like (this is already true by default, without me trying) when I'm around a bunch of Feelers talking about whatever they want to talk about, I just shut up. Except that in a social situation with people just conversing, listening to the sound of their own voice (and not much else), I'm a zero, it doesn't matter if I say nothing; when I'm responsible for getting product out the door, or for teaching young minds how to think so they can program a computer, I'm responsible, and if people don't want to listen, or if they are offended at the possibility that I might know what I'm saying, I can't do my job. I'm geting paid to do this, so morally I need to do it -- even if nobody else believes in doing what you are paid for, God does (and nobody else counts). So I need to spend the extra time and energy on producing an audit trail. The result will come out slower, but God doesn't seem to be in a big hurry the way humans are, so I guess it doesn't matter. The director doesn't complain about slowdowns, he mostly has nothing to compare it to.

An audit trail is insufficient. If one or both parties in a dispute has compelling data (evidence) that the other party does not have access to, they have an unfair advantage. Knowledge is power, but knowledge in my head and nowhere else is a zero. I got myself in deep doo-doo a year ago when a vendor offering "cloud" computing services using bleeding-edge technology, it failed to work on the student computers. I looked at all the data available and blamed them. The director told me not to offer inferences, stick to the public data. They were holding some data close, and I came off looking like an idiot when it came out. He threw that in my face again this week.

Santayana famously said "Those who cannot remember the past are condemned to repeat it." This same director tells me "The past is past, you can't change the past," with the explicit implication that you should say nothing more about it, but it seems that he can't do it himself. Nobody can, if they want to succeed, as Santayana reminds us. The Biblical message is similar, but not so succinct or pithy. The Bible also teaches forgiveness, but not without repentance (promising not to repeat it), which if credible, makes Santayana's line irrelevant.

The Golden Rule is still an excellent way to understand these things. I need to raise my threshold of confidence (at least in public) so as not to make claims that fail to rise above reasonable doubt, based solely on public evidence. I need to remember the past -- not with any silly desire to change it -- but rather to change me so I don't do such a stupid thing again, or if it's something somebody else did, so I don't get myself into that position again. I guess that still is a case of changing me, neither of which needs any public reminder (most of the time, unless they insist).

As for getting my work done when I'm the responsible party, God again has the best answer: I am responsible only for what God has given me the resources to do, and God can take care of the rest in His time. It took some effort, but I managed to persuade the director to let me do what God made me good at -- and the director freely admits that I am -- and he now seems willing to take his own responsibility for what we both know and agree he's much better at, without trying to delegate those tasks to me. Well, he still tries, he's a manager, that's what managers do, so I need to be prepared to push back, perhaps remind him of his promise.

A big part of the problem is that I make a lot of mistakes, which additional thinking can correct. I do that, which makes me pretty good at programming, and sort-of good at preparing teaching materials and teaching from that preparation, but really lousy at on-the-fly interaction, because there's no backspace key, no Undo menu, you say the wrong thing and Kaboom! Conversation over. Relationship over. Mostly the fact that I think slow keeps me out of trouble: the conversation has moved on before I formulate what to say, so I just say nothing. It gives people a really bogus idea of who I am, but that's mostly their problem, not mine. The problem comes when I'm supposed to be running the show, and I don't have enough time to correct my mistakes before going public.

Another big part of the problem is that I'm lazy, I need moral certainty to motivate me to get off my duff and get to work. If I back away from that, then I have no purpose, no reason to stop dithering, and the necessary work doesn't happen. Maybe that's why my productivity drops so precipitously when these social conflicts happen.

The director is a Thinker, but he knows how to manage Feelers, and as part of that he acts as if everybody is a Feeler and smothers me with affirmations, which because we all live in Athens (see Diogenes above) is wasted on me but there's no need to disaffirm him by saying so. My response to him was slightly less than superlative and he feigned insult, so I told him that the best guy I worked with was an engineer, and we worked well together because we each respected what the other was good at. After hearing that, the director agreed to not continue pushing onto me managerial issues that I handle so poorly. It could have come out a lot worse, but this is a win-win relationship, and we both know it and want to keep it that way.

One management skill I might be able to learn from him: I am not responsible for the work of anybody but myself at this time, so I can with impunity treat everybody like Feelers, and never say nor think anything negative about them, and if they don't perform at what we (this organization) need, it's the director's problem, not mine. And any such complaint from me needs to exceed "reasonable doubt" with public data. He is not interested in what I believe or think I know, only what I can support with public data. It might take some effort a few times to resist the temptation to rely on my interior clock, but I probably can do that.

Remember the past. Don't be that guy (in this case "that guy" was me, last year).
 

2023 February 10 -- Management by Delegation

It's not exactly a joke -- to us techies, it's more of a wry observation on how the world works -- I sometimes describe the world as made up of two kinds of people: Those who get their jollies out of altering the positions of the molecules on the face of the earth, and those who get their jollies out of telling other people to alter the positions of the molecules on the face of the earth; the first have lots of fun and the second make lots of money. I'm clearly in the first group and not ashamed of it, it's what God made me good at.

Managers are in the second group. I've been there, but I only succeed when the people I manage actually want to do what I'm telling them to do. Teaching is like that: I cannot teach anything to anybody who does not want to learn it. Maybe that's true of all teachers, but some of them successfully persuade the students that they want to learn this. I can't do that (see "Audit Trail" last year). I'm there again in a second capacity, managing the mentors who are assigned to assist me in the teaching of these students. These are college kids who (presumably) want to teach, and who are getting paid to do it (so they want to do it here), I just need to tell them how we want the kids taught, and they (more or less) do it.

Then there is the boring mechanical stuff of any job. In programming that includes getting the files in the right places and telling the compiler what to do with them, stuff like that. Managers have their counterpart, scheduling the people you manage for their duties, ordering office supplies, stuff like that. It's what executives have secretaries for and army officers have corporals for. Managers don't do the stuff that needs doing, they manage by delegation. What happens if there's no budget for secretaries and corporals? You get the lowest-paid person in the organization to do those flunky jobs. In this case it's the manager himself, he's the money guy, nobody pays him anything at all for this, but that doesn't count. So I guess I'm the stuckee.

When he asks me to do the managerial flunky work that he could do himself faster and more efficiently, he is telling me that he does not value my contribution to the enterprise. I'm a zero, I expected that message to come when it came time to divide up the pie; I did not expect it while the success of the task before us rests so heavily in my hands. Maybe I need to say that to him, but I don't exactly know how.
 

Links

Complete Blog Index
Itty Bitty Computers home page