Share and discuss this blog

Thursday, April 28, 2016

Could IBM stop lying about Watson already? I guess not

IBM needs to stop lying. It is getting hard to take. Today alone I saw two outrageous lies about how Watson will save us all.

Here is the first article:

Its headline is:

Big Data: Will We Soon No Longer Need Data Scientists?

As you can guess, the answer is we won’t because Watson.

IBM, for example, believes that it can offer a solution to the skills shortage in big data by cutting out the data scientists entirely and replacing (or supplementing) them with its Watson natural language analytics platform.

I want to keep this simple, so I will say what I was doing today. I didn’t sleep well last night because of a phenomenon called alcohol rebound. I only had 2 drinks, but I had them 2 hours before bedtime  and this caused a rebound at 2 am which kept me up for hours. This has only started to happen to me in the last year or two, so I Googled “alcohol rebound in old people”  and found a long list of articles none of which were any help. I could ask my doctor but I am guessing he hasn’t memorized the literature and doesn’t know the data. But Watson can do it right? Watson wouldn't even understand my question much less my needs and it would not be able to extrapolate from data that might or might not be there. To put this another way, Google can’t answer most of the questions I pose to it and Watson is no better.  Natural language processing is not very good yet, no matter what all the “AI” deep learning people say. Intelligent people are always better to talk to than any AI system we currently envision.

These days we have large life insurance company as a client for one of our data analytics courses. So I imagined a  question they might ask Watson. “What is is the worst policy we could write?” They might ask that. Would Watson even know what “worst” meant in this context? Would it understand all the parameters relevant to determining an answer? I assume this company’s data scientists could answer this while it is safe to assume that Watson wouldn’t even know what the question meant. But this doesn’t stop them from advertising more nonsense about Watson. I had had enough.

And then I saw this:

The headline is

“C” is for cognitive learning
IBM and Sesame Street collaborate to create the next generation of tailored learning tools. This new technology venture combines Sesame Street’s expertise in education and storytelling with IBM Watson technologies.

And what piece of brilliance will Watson bring to education?Apparently they are just hopping on the personalized learning bandwagon, which means we will teach the stuff we are making you learn by tutoring you to get better test scores when we see what answers you got wrong. So, Watson will change learning, or maybe not so much. Watson will help kids who can’t read well by seeing what words they have trouble with and helping the kids practice. I have news for IBM. People can already do that. Good parents and teachers always do that. Is IBM’s view of education that all kids will have everything they do analyzed and then shoved at them again in another form because Watson is good at analyzing data? 

The problem in education is simple enough folks. It is boring. It is irrelevant to the interests and needs of most kids. They don’t need to learn classical Greek, or ancient history. They should be encouraged to learn what they want to learn. Could we do something radical and ask kids what they to learn how to do and then them help then learn to do it? We could, but then if we submitted the answer to Watson it wouldn’t understand a thing the kids responded. (What would it do with “I want to be a fireman?”)

(As an aside, people who read me regularly know that I am a terrible typist. Apple’s Pages does automatic spell correct and is very bad at doing that. But today it corrected my misspelling of Watson to Satan on two different occasions. Maybe Apple’s AI is smarter than Watson’s.

Thursday, April 21, 2016

Thank You Indiana for reminding me why the government has no idea what it is doing in education: Knowledge of AI now a requirement for Indiana 8th graders

I was a a professor of Computer Science for 35 years. But, I didn’t learn enough about the subject apparently. I would not be able to pass the new Indiana State standards in computer science for eighth grade.

Here they are: 

Screen Shot 2016-04-15 at 2.19.33 PM

I will now attempt to deal with these questions (which I assume will be in the form of a multiple choice test that signifies nothing other than memorization.)   I will assume, for now, that Indiana really wants answers, so here I go:

6-8 CD1: (demonstrate an understanding of the relationship between hardware and software)

Hardware is the box (phone, iPad, Macbook). Software is the stuff you type and the things you click on. 

Is that the answer Indiana? If it is, then all you are doing is teaching the names of something kids already know about. If it isn’t and you want some more complex answer, you will be out of luck, and are engaging in a pointless exercise.

6.8 CD2: (identify routine hardware and software problems that occur daily)

Sorry but I don’t know what this question is asking. Are they trying to teach that sometimes you need to re-boot your machine? Otherwise I have no answer.

6.8 CD3: (describe major component of computer systems and networks)

Sorry Indiana, I can’t answer that question. Why not? Because I have no idea what it is about. Is “router” one of the answers? How about “printer”? I haven’t a clue. But I am sure, Indiana, that you can make kids memorize a list of terms and then announce great results about Indiana kids and computer science.

6.8 CD4. (how is machine intelligence different than human intelligence)

This is, of course my favorite question. AI was has been my field since the mid 60s. (For all I know. I might be one the 5 oldest people in AI at this point.) And, Indiana, I cannot answer it. Why not?

Describe what distinguishes human from machines: 

A machine is what I am using to type this. I cannot type on people. I used a machine to make toast this morning. No human I know can make toast. I drove from the airport to my home yesterday. I used this machine called a car. Even it was an AI car it would not confuse me. I know it isn’t human.

The difference between how machines and humans communicate:

Humans talk to each other. Sometime they type to each other. Some computers say stuff to you such as “can’t find file”  or "a new update is available." But they don’t fool me. The machine is not saying this actually. It is displaying something a human wrote when the software (or was it the hardware?) was made that I am using. The machine is not talking to me even if it used a human voice to do this. I am not delusional. Apparently Indiana is.

Siri, chatbots, Watson, and every other so called AI is doing the same thing: giving voices to something a human wrote, or, in extreme cases giving voices to something some software found and making believe that it is talking to you and giving you an answer. This is not machine intelligence. It is a game that various companies are playing to make you think these machine are intelligent. Is that the right answer Indiana?

Describe how computers use models of intelligent behavior?

At least this question isn’t asinine. I have been working on it for more than 50 years. It is an important question. I am willing to believe that there is not a single person in the entire state of Indiana who knows the answer. Wait. I remembered that one my students is a professor at Indiana University. He knows what the answer is: "we haven’t really figured it out yet." Guess they didn’t ask his advice.

Good job Indiana. You have made school even stupider than it already is.

Sunday, April 17, 2016

Former slaves studying Latin and Greek; nothing has changed

I am in the middle of reading a book called “The Black Calhouns,” written by Gail Buckley. It is a story of one African American family starting in the times of slavery and going to the present. I was not reading this book because of my interest in education, but, as often happens to me, I became infuriated by something I read that related to education.

The book says that in October 1870, the Georgia State Legislature provided money to educate “Negroes” at schools set up for this purpose, but there was “widespread belief that this would not work.” So, they held examinations, “overseen by a board from the old slaveholding class.” A previous Georgia governor said: “I know these Negroes. Some of these pupils were my slaves. I know that they can acquire the rudiments of an education, but they cannot go beyond. The are an inferior race, and for that reason, we had a right to hold them as slaves, and I mean to attend these examinations to prove that we are right.”

After the examinations, the Atlanta Constitution wrote: “we are not prepared to believe what we witnessed:  To see colored boys and girls fourteen and eighteen years of age, reading Greek and Latin, and demonstrating correctly problems in Algebra and Geometry, and seemingly understanding what they demonstrated appears almost wonderful.”

I was taken aback by this since I wasn’t really thinking about the idea that what upsets me most about education has been going on that long. They were teaching the newly freed slaves to read Latin and Greek and to do Algebra. Why?

If you asked me to design a curriculum for these children it would have had two main principles. First, it would have offered choices. I have never understood why every child must learn the same stuff. Second, the choices would relate to the real possibilities of the future lives these children faced. Were these kids going to become scholars in the Classics? Were they going to ever use algebra for any reason?  I would have taught them how to open a business, how to run their own farm, how to fight for their political and economic rights, how to think critically about life decisions they might actually have to make, how to become articulate, how to get along.

I hadn’t realized that today’s silliness was going on in those days as well. Today, for example, in New York City, there is a charter school that seems to be everywhere with lots of funding, called Success Academy. When you look at their website the faces of kids that they show are almost all non-white. The curriculum that they offer might as well be the one offered in 1870 to the former slave children. It is the same nonsense.

What was going on then, and what is going on now, is the attempt to prove that these kids can go to Harvard and become scholars and Supreme Court Justices. I am sure that some of them can. But how many? One percent of them? Not that many even.

We have held the collective insane belief, and now I realize that this belief has been around a long time, that the way we help poor children to live better lives is to treat them as if they were very wealthy children who may not actually ever have to work and for whom the world is wide open to them.

Poor children should be treated the same as rich children. Sounds good. Sounds democratic. A lovely ideal. Because we want to believe this, we have closed up vocational schools and made education all about preparing for college.

Let me remind the people who do this, that going to college is just as likely to leave a student in massive debt and with no ability to work because he was convinced to become a literature major.

Even in 1870 we were preparing children to be scholars. Why were they learning Latin and Greek? The answer was that all the “important books” were written in Latin and Greek, but that was never the real answer. Even in 1870 there were books written in English. And, although we don’t make every child learn Latin and Greek any more, we do still make every child algebra. (And, I might add that my daughter was made to learn Latin, so this still goes on.)

The time has come to get over this nonsense. We can offer hundreds of choices and let kids decide how they want to proceed. The argument against this has always been “but if we don’t expose them to Chemistry, how will they know if they like it?” How many chemists are there? Must we expose every kid to every scholarly field? All it does is create trouble. I was “exposed” to mathematics for sixteen years in school. I liked it. But it was a complete waste of time. When I learned what mathematicians actually did all day, I realized that this profession made no sense for me. But I was never taught that and so I kept studying it because I liked it.

It is time to let kids know what job options exist for them and help them make good choices while also teaching to think hard, make life decisions, learn to speak and write effectively, and generally learn how the world around them works.

I have no information on this, but I am pretty sure that the former slave kids did not go on to be scholars. Neither will any more than 1% of the graduates of Success Academy. There really aren’t that many jobs for scholars.

It is time to become realistic about what we teach in school. We can offer a scholar track too but people need to know what scholars do all day and how may jobs there are for scholars. We simply have to stop being stupid about education.

Wednesday, April 13, 2016

A history test from AP; test your ability to stay awake

Below is an article from Sunday’s New York Times Education Supplement. I am simply posting it here. My point is simple enough. Why do we have tests like this? Whose interests do they serve? Who remembers what is “taught” by them?  And, how do they possibly relate to how a student will do in college? (Actually that last one I can answer: college is full of tests like this as well, at least bad college course are.)  No wonder students are bored to death in school and can't remember what they "learned."

U.S. History, Revised
Roundly drubbed as left-wing anti-Americanism, the framework for the Advanced Placement course in United States history was recast for 2015-16. Here are some of the practice questions that were revised to address issues.  

Refer to these quotes when answering questions 1 to 3.
1The statements of both Truman and Reagan share the same goal of ...
restraining communist military power and ideological influence.
creating alliances with recently decolonized nations.
re-establishing the principle of isolationism.
avoiding a military confrontation with the Soviet Union.
2Truman issued the doctrine primarily to ...
support decolonization in Asia and Africa.
support U.S. allies in Latin America.
protect U.S. interests in the Middle East.
bolster non-communist nations, particularly in Europe.
3Reagan’s speech best reflects which of the following developments in U.S. foreign policy?
Caution resulting from earlier setbacks in international affairs.
Assertions of U.S. opposition to communism.
The expansion of peacekeeping efforts.
The pursuit of free trade worldwide.
Adolph Treidler/Collection of Library of Congress
Refer to this image when answering questions 4 to 6.
4The poster was intended to ...
persuade women to enlist in the military.
promote the ideals of republican motherhood.
advocate for the elimination of sex discrimination in employment.
convince women that they had an essential role in the war effort.
5The poster most directly reflects the ...
wartime mobilization of U.S. society.
emergence of the United States as a leading world power.
expanded access to consumer goods during wartime.
wartime repression of civil liberties.
6Which of the following represents a later example of the change highlighted in the poster?
Feminist challenges to sexual norms in the 1970s.
The growing protests against U.S. military engagements abroad in the 1970s.
The increasing inability of the manufacturing sector to create jobs for women in the 1970s and 1980s.
The growing popular consensus about appropriate women’s roles in the 1980s and 1990s.
Jacob A. Riis/Bettmann, via Corbis
Refer to this image when answering questions 7 to 9.
7Conditions like those shown in the image at right contributed most directly to which of the following?
The passage of laws restricting immigration to the United States.
An increase in Progressive reform activity.
A decline in efforts to Americanize immigrants.
The weakening of labor unions such as the American Federation of Labor.
8The conditions shown in the image depict which of the following trends in the late 19th century?
The growing gap between rich and poor.
The rise of the settlement house and Populist movements.
Increased corruption in urban politics.
The migration of African-Americans to the North in the late 19th century.
9Advocates for individuals such as those shown in the image would have most likely agreed with which of the following perspectives?
The Supreme Court’s decision in Plessy v. Ferguson was justified.
Capitalism, free of government regulation, would improve social conditions.
Both wealth and poverty are the products of natural selection.
Government should act to eliminate the worst abuses of industrial society.
Refer to this quote when answering questions 10 to 12.
10Which of the following aspects of Muir’s description expresses a major change in Americans’ views of the natural environment?
The idea that wilderness areas are worthy subjects for artistic works.
The idea that wilderness areas serve as evidence of divine creation.
The idea that government should preserve wilderness areas in a natural state.
The idea that mountainous scenery is more picturesque and beautiful than flat terrain.
11 Muir’s ideas are most directly a reaction to the ...
increasing usage and exploitation of western landscapes.
increase in urban populations, including immigrant workers attracted by a growing industrial economy.
westward migration of groups seeking religious refuge.
opening of a new frontier in recently annexed territory.
12Muir’s position regarding wilderness was most strongly supported by which of the following?
Members of the Populist movement.
Urban political bosses.
American Indians living on reservations.

Preservationists concerned about overuse of natural resources.

Wednesday, March 30, 2016

Pragmatic Learning: It's not "fun"

In the mid-nineties Microsoft invited me to see what they were doing in education. I looked at what they had built and laughed. They had animated some chemical equations. They thought that would make chemistry more interesting to kids. (They never asked why it mattered that kids balance chemical equations but knew that if they were animated the experience would be much better somehow.)  

I was reminded of this the other day when a pharmaceutical company asked us to do work for them and insisted that whatever we built had to include animation. I asked if they wanted to show a whale eating a planet or something that one could only show through animation, but they said “no, they need their training to be fun.” Of course, this animation thing is just part of a larger problem. Most training is boring.

In general, training is not fun. Actually, it is quite unusual when any formal learning is fun. When I think about learning and adjectives to describe good education, I think of profound, exciting, insightful, thought-provoking, but not “fun.” Are games fun? This is an important question for people in training because not only animation but now “gamification” is a new trend. But are “games” fun? Winning is fun. Interacting with others with whom you are playing can be fun. Games can be entertaining and sometime they are fun, but when we think about making training more effective, we need to think less about having fun and more about what it means to learn. These are odd ideas I know, but actually very important ones, so let me explain:

I play softball. Learning to hit involves many years of trial and error. I have been playing softball for more than 60 years. I am not an expert at hitting. I am not bad, but there are always people better then me. When I happen to hit a ball so well that it gets by everyone and I can get a triple or a home run, I am very excited. I am pleased with myself. People cheer for me. Is it fun? No.

It was the result of hard work, and I am psychologically invested in doing it. But as any softball player will admit, the fun part is a “come from behind” win in the last inning or laughing about the game with friends over beers later on.  Hanging out with friends and laughing is fun. I don't learn much in those circumstances however. When I learn, it is because I have a goal and have worked hard towards achieving it. It is satisfying to succeed. Of course, I enjoy whatever victory might come, but fun? NO. It is the wrong word and the wrong idea. We are confused about what learning is really about. So we create silly words, like nano- learning, or make absurd references to neuroscience and lose the forest for the trees.

When parents raise children before they go to school at 5 or 6, what do they teach them? Nothing that looks like something in a course or a classroom.  

Parents don’t put kids in classes or courses when they are little (unless, of course, they need some day care.) They do not sit them down for lessons. Nevertheless, in early childhood, children learn to speak their language, navigate their house and their neighborhood, get along with other children, operate within the family rules and structure, and they learn whatever might be of interest to them from how to play with dolls or trucks to how build a city with blocks. No lessons. No courses. 

Teaching? Mostly it occurs when they need help. My daughter said to me after coming upstairs to ask me a question one day when she was 5, “I will be back when I need you again.” (She has been coming back ever since.)

My purpose in writing all this to make a clear a controversial point. We need to stop thinking that delivery of learning is about creating “courses.”

Really? Always? No more courses ever? How will we train people? 

I don’t mean that we should never build courses, let me make that clear. So, let me start with which kind of courses should be saved and which should go. I started helping Andersen Consulting (now Accenture) with learning and training in 1989. They knew what they had wasn’t very good. They made trainees read “green books” (referred to by one and all as FGBs.) They made people attend courses who were clearly sleeping through them. They knew they needed help. They told me that they would choose to offer a course on something in say March and that the attendees would include people who had been working on that thing for six months already and some who wouldn’t be working on that thing for six more months. But they could only hold the course once a year. They figured that building an online course might improve the situation, but this kind of issue was endemic at Andersen and speaks to the first real problem in making people take courses.

The people who take the course may not need what they are being taught at exactly the moment the course is offered.

In college this is nearly always the case. We are so used to it that we expect when we take a course that we may not use what we have learned for years or ever. We make kids take algebra because “they will need it later" when hardly anyone ever does. Corporate training people ought to be smarter than that, but oddly they are not. (They all went to school so they think training should be like what they know.)

Why does this matter? Let’s think about children again. We wouldn’t offer a course about “the past tense in English” to our child who just said “taked” instead of “took.” We would simply correct them. Courses are often offered because companies can’t teach at the exact moment of need. Well, I am here to say that they can. One issue in improving training is to convert courses offered every now and then into experiences that include “just in time help”. This is very important because people forget what you teach them when they can’t immediately put it into use. A child will say “took” as soon as they are corrected. An employee will not remember what they were taught if they can’t use that information immediately. 

This leads us to a simple idea:

Eliminate the majority of courses and replace them with experiences that contain just in time help when that is possible.

Now let’s think about other kinds of courses.

There are a many courses that attempt to teach the impossible. “Leadership” courses are a good example of this phenomenon. How exactly would you learn to lead from a course? Should we talk about the ten principles of leadership?  Should we read books on leadership and discuss? All the talk in the world does not make someone into a leader. But companies do have this problem. They want to train leaders. What should they do? If they stop thinking that they want a course, it would help.

How do children learn to be leaders? All through childhood there are kids who tell people what to do and there are kids who listen to them. How does this happen? It happens very simply, actually. It happens naturally. Some kids want to lead and some kids want to follow. Some kids want to lead but no one listens to them. Others lead and are followed.  So, I am skeptical about leadership courses. On the other hand, managing a project is complicated and it would be a lot better to manage a few fictional projects in fictional situations than it would be to learn project management on the job and possibly screw up something important. Leaders do learn to lead better over time. Project managers learn to manage projects better over time. The difference is between what I will call natural skills and artificial skills.

Speaking is a natural skill. Some people are good at it and others aren’t. Anyone can get better at it over time, but I wouldn’t be a big fan of a “how to speak” course. Having someone who criticizes speeches you give is a different story however. When I was just starting out, someone I respected said to me, right after he heard a speech I gave: if you try to say everything you know in a hour, Roger, you don’t know much. I didn’t need to take a course in speaking. I needed to be critiqued just in time. As you can see, I still remember that lesson. It wouldn’t have meant much if I read it, or heard it in a lecture, or in course, but because his advice was about me and what I had just done, it stuck with me. This is how we learn — not through courses but through experiences. And, that experience is much better understood when we have someone giving us some good advice about what we just did.

So, another problem in building courses is this:

Don’t build courses that attempt to teach something that no one has ever learned from a course in real life.

Then what should we do? To foster leaders, an organization needs to look and see who is leading. Then the job of the training organization is to help people who are natural leaders lead better.  But sometimes we need our employees to do something that might not be entirely natural to them. What do we do then? We must build a situation that they can try out and help them get better at it. You can call this a course if you like, but it really is something quite different from what passes as a course in school or in training.  

There are courses that are worth building. These always have the same property. Everyone is on the same page at the same time. If an airplane manufacturer needs to teach people how to operate or repair a new piece of machinery, a course is just what they need. It should be a learn-by-doing course with lots of practice and just in time help. No one will learn to do this without a course, and individual instruction is not important to focus on when many students are in exactly the same situation and all can be handled simultaneously. Notice this means they can all practice immediately. This is what learning by doing is all about.

But, and this is an important point, this does not justify building algebra courses, or chemistry courses, or history courses. No student needs to learn algebra at a particular time. There could be a need to learn certain aspects of mathematics within the context of doing something that requires it. A short course in, say, an aspect of statistics to help someone understand how to interpret data who is actually needing to interpret data is the right kind of mathematics course. We have gotten caught up in the school model of courses where everyone has to take a course whether they are interested in learning the material being taught or not and the material must completely cover the subject area.  The fact that this goes on in corporations is nothing short of insane. Schools provide courses because the structure of the school has only so many teachers, students who need to be kept busy all day, and government regulators who like to make rules and tests.  Corporations do not have this problem, especially when the material can be offered online.   

Courses need not be administered to multitudes. One can have a course that is for one person only and can be used when needed. Such a course must be online since we can’t expect teachers to show up just when you need them. The reason to build an online course is not so you can have 10,000 attendees.   There are two very important reasons to build online courses however. The first is that is possible to do things in simulation that are not possible in a classroom. Air flight simulators are a very good way to learn to fly, for example. We need simulators for everything. These need to be on computers so people can practice all kinds of skills when they want to (or when their company needs them to). The second reason is that teachers are very important for learning, not ones who lecture you, but ones who notice what you just did and can give helpful hints or answer questions. In an online world these teachers can be readily available, If you want to design an airplane, the beat teachers may be in Seattle. It just shouldn't matter where you are. Online courses that contain simulations and give one the opportunity to try things out, learn from one’s mistakes, and practice, are the future of education in school and at work.

We can also build simulations using no computer at all. We can create simulated experiences amongst a group of people led by experts who create realistic situations and help trainees profit from those situations. The computer might very well be irrelevant. The real issue is having real experiences, and conversations about those experiences directed by an expert. 

Just doing a course all by yourself may not be the best idea. We do need help when we are learning and we do need people with whom we can discuss new ideas or problems. So, we can, and should, build courses for people who will take them when they want to, but we must provide, and this is not hard to do in an online world, connections to other people who are taking the same course, so that ideas and lessons learned can be discussed. We don’t really learn without practice and part of practice is conversation. Practice includes talking about what we are thinking.

The real issue having an appropriate vision of the online world in which education and training will eventually completely reside. Classrooms will disappear. You Tube, and TED talks, which look like the kind of replacement I am talking about, are still full of talking heads. MOOCs are still classes and lectures but without the physical room. 

So we need courses it seems, but really do we? What could we have instead? Let’s return to thinking about small children. What children have, if they are lucky, is a parent who is always around. Some kids are sent off to day care as fast as a place can be found for them. Then, they are in classroom- like situations all their childhood and are always treated as part of a mass. My main problem with courses is, of course, exactly that: Massification. This has become one of the “in” buzz words in the training world, sending exactly the wrong message. Little kids who do have a parent around also have toys, games, and trips to the park, or zoo, or store, or parties. In other words, their parents provide them with experiences, and it is within those experiences that they have questions and can initiate conversations and get help.

What this tells us is that real autonomous, motivated, learning happens when you are in the middle of doing something, and questions arise in your mind about it. This is exactly what we need to build into corporate training (and into school if they could possibly change their models.)

I learned what I have just said from an experience (of course). In this case, we were building sales training for one of the phone companies. They were selling Yellow Page ads. We built a learn by doing course, going through many different experiences and issues. But the sales trainees were smarter than we were. When they were about to sell an ad to a doctor, they found the “selling a yellow page ad to a doctor” part of the course we had built and took it before they called on the doctor. When they were selling to florists  (who behave very differently than doctors,)  they took the florist part of the course. My team learned that we could build training in pieces, meant to be delivered just in time. 

We learned this again when we built a coaching course for IBM. People weren’t going through the entire course. They would use it when they anticipated a difficult coaching session and would find a similar scenario in the course to go through prior to their actual coaching session. You might well forget what you learned just in time, but that would be fine because you could always practice it again. One would assume that after a while one wouldn’t need to keep re-learning, but what would be the harm in brushing up on the way to making a sales call or coaching session?

Would that be fun? Suddenly fun is important in the training world as is “bite-sized nano learning", “gamification,” “badges” and many ways of all saying the same thing: Most people think that doing training is boring. And of course, they are right. But the opposite of boring is not “fun” or a “game” or “nano.”

To understand this we must think more about fun and think more about learning. I used to ask my undergraduate students, (just for fun) to tell me what they had learned that day. I never heard anyone (not even once) respond with something they had learned in a classroom. They had learned something about their friends, or about life, or about themselves. Course work was never mentioned. The stuff they told me was never fun stuff. It might be that they had learned what their girlfriend needed from them, or that they shouldn’t order the hamburger in the cafeteria ever again.  On a daily basis we learn a lot about the world we live in. It is rarely “fun” to learn it, but is often important to learn it. 

What does this tell us about training? If you have to put something in a game format in order to make someone learn it, you are teaching the wrong stuff in the wrong way. If people won’t take your course because it doesn’t have fun animated characters, it is because the material is boring, and more importantly, because the trainee has deemed it to be irrelevant to his or her success.

On the other hand, it is inherently interesting to learn something that you think you can put to immediate use. We need to understand that “interesting” and “useful" are not the same as “fun” and “badges.” We need to make sure the training itself is interesting and useful, and not worry about the trappings we surround the training with.

So how do we do that?

Let’s think about small children again. What do they need in order to learn? 

1. something they are trying to do
2. someone to ask when they need help

It is really just that simple.

In order to make learning in childhood work, we create or enable situations that are interesting or appeal to some intrinsic goal (like eating). And then we make sure help is available. We also enable discussion and approval: (“Look at what I just did. Did I do it right?”)

With this simple idea, I have told you all you need to know about training. Because if we can build “on-demand online training,” we can change the world of training significantly. This is what we have to do:

  1. We need to anticipate the needs of trainees
  2. We need to provide a way for them to satisfy those needs
  3. We need to provide people for them to discuss things with

This means that we should have:

1 experiences to try out virtually, available on demand

2. online mentors, available on demand

3. co-workers with whom to discuss experiences

Can we do this? Of course. We need to stop building courses and provide an over the shoulder autonomous entity that knows what you are working on and can offer help. That help would range from just-in-time advice, to just-in-time practice in a new environment, to more prolonged course-like material when there is something complex to learn how to do. This is what important AI would look like. We should not have courses that provide information. We should have courses that provide experiences. And we need to provide mentors and peers with whom to have conversations. Those mentors should be people until we can build good AI mentors.