Working Up

by Dwayne Phillips

Agile thought: do, learn, change, do learn, repeat. Great stuff in some situations. Not great in national policy.

Someone recently noted that the Agile Manifesto was 20 years old. This was a fancy way of stating the obvious when it comes to experiments: do a little, learn a little, do a little, learn a little, etc.

The computer as we know it today is perfect for this “agile” idea. Type words on the screen. Save them, read them, improve them, repeat. Type a computer program on the screen, run it, learn, change it, run it learn, repeat.

National policy. Sorry. That is not like writing something on a computer screen. Do a little, learn a little, change, do a little, learn a little, repeat. It doesn’t work with national policy about health or anything else at that level. Spend a few billion dollars. Learn a little. Repeat. Nope.

“We know more now” is a pathetic excuse for lack of thought.

An agile nation? Nope. Won’t work. Now, let me read the words I types, change, learn, repeat.

by Dwayne Phillips

All this mask wearing allows me to not recognize and not talk to people when I’d rather just keep doing things I’d rather just keep doing.

The year of the virus has brought us many changes. Let’s consider one that I discovered recently and like (on those occasions when I like it).

I find it difficult to recognize people when they are wearing a mask. Some people find it difficult to recognize me when I am wearing a mask. Since I live in one of the enlightened districts of America (that is both a compliment and a disparagement), everyone is still wearing masks here in mid-April 2021.

Situation: I am walking the sidewalk of a shopping center (you know the common type that is anchored by a grocery supermarket). I see someone I know. I recognize them despite the mask. By the glance of their eyes and the raising of their eyelids, I recognize that they recognize me.

I don’t want to stop and talk to them in this place at this time. Ah, here is the benefit of the mask. I just keep walking.

Sometimes we meet the next day in a situation where talking is either welcomed or unavoidable.

“Hey, were you at the shopping center yesterday?” the other person asks.

“Well, yes I was,” I answer.

“I thought I saw you, but you kept on walking so I thought it wasn’t you,” they say.

“Oh,” I reply. “I didn’t notice.”

“Yes,” they say, “with these masks sometimes we just don’t recognize people we know.”

“I know what you mean,” I say.

See? The masks give us anonymity at times when we just want to keep doing what we want to keep doing. Perhaps this is selfish. Perhaps this is finding some good in an otherwise bad situation—and I believe this pandemic year is a bad situation. I find that all the mask wearing is bringing more harm than good, but that is just me. I find it a good excuse to not recognize, stop, and talk to some persons in some situations at some times.

by Dwayne Phillips

Another variation of Lewis Carroll’s “If you don’t know where you are going, any road will get you there.” This time we apply it to hiring data scientists.

“We can’t hire enough data scientists.” At least I read that almost everyday out there on the Internet. Well, why can’t we? Here are a few of my reasons.

(Reason 1) We keep raising the bar. It is not enough to (a) know programming, (b) know math, science, and statistics, and (c) know a subject area. Now we need to know artificial intelligence and machine learning as well.

“If you cannot deploy six different regression models on AWS SageMaker in less than 15 minutes, you are not a data scientist.” Or so the thinking goes.

(Reason 2) We keep lowering the bar. We want to “democratize” data science so that you drag a data file into a GUI, type a question, and hit Return.

(Reason 3) I guess we can’t agree on a definition of a data scientist. Hiring an Electrical Engineer is much easier; look at the college diploma. We don’t have many folks running around with a diploma that says “data science.” Hence, we have to think a while before hiring, and thinking gives us a headache.

We have the unknowing trying to find the undefined. Any road will get us there or get us lost in the woods. The results are predictable and predicted.

by Dwayne Phillips

Sometimes decisions just sort of drift in and out with the tides.

The tide comes in; the tide goes out. Sometimes we find something that drifted in with the tide and remains on the beach. The next day, the next cycle or two of tides, and that something is gone. It was carried back out by the tide.

I have seen managers decide important questions like this. One day, they wake and, well, there you have it. Decided. A few days or weeks later and it is gone. Decided.

Nothing of consequence happened before each decision. The decisions just sort of happened. Just like the tides.

We often do just fine by considering the beach and what has come in or out with the tides. There is enough momentum and expertise in all the folks surrounding us that our decisions neither help nor hurt that much.

Sometimes are decisions shouldn’t be left up to what the tide brings or takes. And we shouldn’t let the tide decide if we are deciding by the tide.

by Dwayne Phillips

We just cannot resist applying the latest technology to something.

This is the second in my series on the “Technology Imperative.” (I have learned that there are books and papers with this title elsewhere.) The first part in the “series” was blogged in the year 2010. Oooops, not much continuity, but here goes.

Our American government scans faces at border entry points to find imposters, those folks whose desire to enter America leads them to break the rules. We find this story of how 23 million faces were scanned in 2020. All this technology resulted in the arrest of, well, uh, er—nobody.

One could claim that this is a tremendous success as the use of this technology prevented anyone from attempting entry. One could claim that this was a big waste of money since no one was trying to enter. I guess there is no end to how others could claim other things.

Nevertheless, the technology exists, so we must use it. That is the technology imperative. There appears to be no end to very expensive examples of this. Sigh. What can we say about ourselves? Hey, wait, look at that shiny new gizmo. Why we could…

by Dwayne Phillips

Despite all the promise and all the already fulfilled promise, there is a fundamental problem with machine learning.

Machine learning is that part of artificial intelligence that time and technology favors. We have the stuff needed to make machine learning work right now.

Machine learning “works” because you feed in a million photos of things (at least for computer vision, the other ML areas have the same situation), learn, i.e., create a model, and then use it.

The technique can “read” x-rays, find cracks in airplane engines, warn folks of impending forest fires and all sorts of things that benefit people.

Then there are those applications where the technique finds a criminal in a crowded room by recognizing a face or points to someone who “doesn’t belong.” We find that there are all sorts of “biases” in those million photos that some of us don’t like in some situations. What do we do?

Solution: go through those million photos by hand to ensure we have a set that is “fair” or “better” in some way. Going through all those photos by hand defeats the purpose of the entire exercise. We will be inspecting everything manually to ensure that the machine does everything automatically. Huh? Won’t work.

Well, we can still justify this as you have to look at a million photos once to create a good “model.” Afterwards, you use the model over and over and over. That reuse has a greater return on investment than having people do all the work from now on. Really? Have you run the numbers on that? Well, uh, …

There is something some of us call “the Technology Imperative.” It states that if the technology and technique are available, we must use them. That usually wins the day.

Perhaps something good will come of this.

by Dwayne Phillips

Where will we go for discourse on the Internet? Semi-public groups for forums work.

I like Seth Godin’s recent short essay on trust and how folks use the Internet. Godin described the early (1970s) online interaction as, “Because each of these groups were high-trust communities, it was easy to conclude that the people they’d be engaging online would be too.”

Godin mentions the unfortunate online interaction of this century, “When a site decides to get big fast, they usually do it by creating a very easy way to join, and they create few barriers to a drive-by anonymous experience. And when they make a profit from this behavior, they do it more. In fact, they amplify it.”

Money is the root of much of this. Open your platform to everyone, increase traffic, increase ad revenue, get rich.

Godin points to the future and a better way, “Until there’s a correlation between what’s popular or profitable and what’s useful, we’re all going to be paying the price.”

Another solution: start your own little invitee-only discussion forum that is not hosted by the big and famous tech companies. Jerry Weinberg’s “SHAPE Forum” was one I used for a decade (This link goes to a sample discussion. In the sample, names were replaced with generic ones.) People who knew one another and understood the language of that group discussed complex matters of importance. Disagreements were common as were suggestions and solutions.

I describe the SHAPE Forum as a semi-public one. It is open to the public as long as the group members invite someone and explain the local guidelines for discussion. Persons who violate the guidelines are removed. If you are removed and don’t like it, well, no one violated your civil rights be allowing you in or moving you out.

Create a forum for your extended family. Family news is shown. Create a forum for persons engaged in the same industry or trade as you. These things are fairly easy to do.

Why do this? We are scattered across the country and the world and time zones. We know smart people who have answers to our questions. Discourse helps us all.

They key is to avoid hundreds of millions of dollars being involved. That is pretty easy to avoid. As long as the money isn’t worth the effort, no one will bother you.

by Dwayne Phillips

We usually build things for ourselves. We then hope that someone else will use what we build. Sometimes we adapt to doing it right; sometimes not.

A recent story related how we want 80-somethings to register for and then receive the virus vaccine. “All you have to do” is go online and… Wait, is there a contradiction here? “80-somethings” and “go online” seem to not go together or something.

A 30-something built something for 30-somethings. Then someone told 80-somethings to use this 30-something gizmo. Hmmm. Sort of doesn’t make sense, huh? How can smart people do this?

Consider Facebook. It was built by college students for college students. Remember how you needed a dot-edu email address to get an account?

Those college-student builders received and heeded some good advice from somewhere. They revised the software so that other folks (not college students) would use it. Viola’! A trillion dollars or something in profits.

Back in 2008, I visited a company that was under contract to build gadgets for the 2010 census. Someone at the Census Bureau thought it a great idea for census takers—those who walked neighborhoods and knocked on doors—to have a little digital device in their hands on which to enter information that would be zapped up to a computer where wonderfulness would occur. Oh, by the way, those census takers were usually retired volunteers in their 70s who had poor eyesight and had just mastered the push-button telephone at home. A 30-something building a gizmo for a 70-something.

I guess we don’t learn fast or maybe some folks learn faster than others. And then there are those who do learn and rake in a trillion dollars.

by Dwayne Phillips

The path to solution is often found in the next (logical) question.

Person A: How much will this cost?

Person B: I don’t know?

Person A: When will you know?

Person B: Uh, I don’t know that either?

Person A: What will you need to do so that you know?

Person B: I’m not sure. I’ll to think about that.

Person A: When will you sit and think about that?

Person B: I don’t know.

Person A: Are you free tomorrow morning at 9 a.m.?

Person B: Yes, I am.

Person A: Then sit tomorrow at 9 a.m. so you can learn what you need to know to know when you will know about how much this costs. That way we will be moving forwards.

…

The next (logical) question is quite simple. If you don’t know something, start tomorrow at learning what you need to know so that you will know something. For some reason, we struggle to ask the next (logical) question of ourselves. We need another person to ask us.

Where is that other person who will ask the next (logical) question? Don’t know? What will you need to do to know? When will you sit and learn what you need to know so that you can…

by Dwayne Phillips

Three years from now may never come.

“I don’t want to be ‘locked into’ company A because company B may have a better offering in three years. We need to be able to shift to company B.”—a well-meaning advisor.

In three years, the world may be different. Let’s remain agile. Let’s keep the ability to move in a different direction.

True, the world may be (probably will be) different in three years. True, it would be good to be able to shift direction then. True, agility is good. True, true, true.

What may not be (and probably will not be) true is that what we know today is relevant to the future of three years from now.

Who predicted 2020 back in 2017? Huh, well, anyone? But 2020 was an extreme case. True, but “extreme” is a relative term, and we will probably think that 2023 was an extreme case.

Hence, while looking to the future is good, it is easy to over rate it. Let’s do something today, now, and gain what we may today, now. Tomorrow has enough worries of its own (I seem to have read that somewhere in a good book).