Steiger is the founding CEO and executive chairman of ProPublica. Engelberg and Tofel are co-CEOs. Engelberg also is editor-in-chief, and Tofel is president.

DL: So you think the teaching hospital is a good idea?

Tofel: I think it’s a very powerful idea but I haven’t seen anyone doing it, and I’ve had long talks with [former Columbia dean] Nick Lemann, who has talked about how they’re doing it, and I think they’re not.

DL: I haven’t seen anyone doing it, not the way [Knight Foundation’s Eric Newton] is talking about it.

Tofel: Here’s why it seems to me they’re not. A teaching hospital is not only a world-class educational institution but also a world-class treating institution. You can have a different kind of thing, but if you are serious about this as a metaphor, you will have to have a school like Columbia, and then you will have to integrate the students somehow into world-class news organizations. To me, the obstacle to this has been whether the journalism schools are prepared to lose some control in order to integrate their students’ experience into the first-rank news organizations. The second problem, if they were, would the first-rank news organizations be receptive?

DL: I think that’s a big question.

Tofel: For us, I can imagine ways in which we would be. Some of the more traditional places may not be.

Steiger: If Columbia physicians and surgeons were in this great teaching hospital, they don’t send all their patients out to clinics downtown. They do a lot of the work on site because they have a lot of physicians there. If you go back to journalism education 20 years ago, it was a dreary hellhole. First of all, if they wanted to do stuff that got published so that ordinary human beings would see it, the newspapers wouldn’t publish it and the magazines mostly wouldn’t publish it because they had all the staff they needed and they didn’t want any help. They embraced the “those who can, do, and those who can’t, teach” kind of philosophy. What people in journalism education were required to do, besides teach the kids, was to publish about the history or theory of journalism. We are now in a world where there are no more barriers to entry. If you choose to, you can do whatever quality of journalism you want in the university setting.

Tofel: You have pointed out the missing, unstated link in my argument. I do not believe a university is capable of building a world-class news organization anywhere. They cycle too slowly and are too risk-averse. I think the whole thing isn’t going to work in a university.

Engelberg: Can I add a small footnote? I was just having this same conversation with my neighbors and friends at Montclair State University because they are in the midst of imagining something more ambitious. They are asking people who are in the neighborhood, like half the journalists in New York, to offer opinions. I do think there is an added complexity of being a state university. They are certainly slow and entrepreneurially risk-averse. Even between private and state, there is a difference. State universities are facing a very challenging environment as you go into this thing. I could say to you right now that Hoboken could use a hyperlocal news site. I guess if you start exposing the mayor it would be considered payback. I think it’s a concern that no one has thought about.

Steiger: There are all kinds of challenges, cultural challenges, to make this work. There are also advantages. What do universities have? They have office space. They have computers. They have lawyers. Things you need. I’m defining away the problem. If there were a will to do it, you could take resources that previously were underused. There was underemployment in the journalism education domain. You could have a faculty that was built more on people that come from the cutting edge of the practitioner tract as opposed to the theoretician tract. The other thing, in terms of the analogies, journalism is not as difficult as heart surgery. There are things that bright young people, with the right kind of leadership, certainly not one practitioner with 50 students, but maybe mix …

Tofel: I have to say, if a big part of journalism is comforting the afflicted and afflicting the comfortable, I think the modern American university is peculiarly poor at that. The development office will be against you every step of the way. The legal department will be against you every step of the way.

Engelberg: A slightly different notion that builds on your notion: What would happen if you take our data operation, and imagine that is the core of journalism education? We all know that the greatest need right now is people in the world of journalism who are not innumerate. When you find them, it’s very rare. And we know when you train people in those fields, they will have jobs. If you can imagine a journalism program that is built on some of the principles we’ve said, which is that in a given geographic area, the creation of a statewide database is generally going to be more labor-intensive than any one news organization can do. What if there was a program whose purpose was to create important data about a state? What if the journalistic execution of that involves some combination of students and local journalism institutions?

This is kind of what we’ve pioneered. It’s a great model because each of those projects and data sets requires a different journalistic skill. Each story is going to teach you how to report a story, what’s a fact, how to write a story, how to transform data into something that people actually read. You then get a hybrid of what The Texas Tribune does for Texas and what we do for the nation. It seems like if you combine these skill sets of traditional journalism professors with the new data journalism—programming, numerate thinking—everyone may benefit. We might get a lot more numerate journalists and a lot more young journalists who understand where the boundaries are. That was my small vision for Montclair.

We have some kids like this. I am not saying that everyone that goes to journalism school needs to code. Some people can code, and some can’t. I think everyone needs to understand what the potential of coding is, or the potential of statistics are. If you can’t do it yourself, you need to know how to talk to the people who can. Paul here can do some math, and it’s fun to watch. I could appreciate what he did, and I could read the equation. Not everybody who can write the code can also be a reporter and a writer, but I think putting the two worlds together and creating more of both is a very viable goal for journalism.

Steiger: I think the clearinghouse thing I think you can do in other ways, too.

DL: What does that mean?

Steiger: Provide various kinds of central resources for news organizations.

Engelberg: If you look at what a journalism student needs to know, it seems first and foremost they need to know how to write and what a fact is. Those things, I believe, come from writing and being critiqued. I think that’s the first step. The business about knowing what a reportable fact is is actually not trivial. When we hire young kids from various kinds of backgrounds, I think there’s still some shakiness about that. The very old-school training at small-town newspapers and the cliche “big city editors” don’t exist anymore, and that is a shame. They were useful and foul-mouthed and abusive even; they did not read the “how to manage Millennials” handbook. Recognizing what is a ground-solid fact, and how to write about things, and how to tell a story are all important skill sets. For that reason, you have to have a journalism education program that involves writing and a critique of that writing by competent people.

I think that one of the things that is becoming evident as the Web develops is that there are various streams of ways to acquire information. On one level, the simple question of who won the ballgame last night, and that story can be written by a robot. If you want to find out whether the Yankees should be dumping their players and giving up on the season, you need to read something written by someone with a brain in their head, and they need to know how to present that information. We have a story that I am very proud of reporting. It’s taken a level of expertise and numeracy that is mind-blowing. Now we have to write a draft of the story that meets the full value of what we’ve got and makes anyone care and makes you want to read it. The draft we have now does not. We will be sitting down with many professional journalists and tearing our hair out to improve the writing and presentation. I think in this crowded world, that skill is going to get more important and not less.

I think it’s a tool box. I think if you were building the perfect journalist from scratch—these people don’t exist, but we can dream—you would want a person to be able to take you to the last moments of the Malaysia flight being shot out of the air. You want to see it and hear it and see the drunken Russians and explaining the things the way the old WSJ used to do in a brilliant way.

Steiger: Bill Carley [of the Wall Street Journal] used to take the final report from [the National Transportation Safety Board] and the facts about what caused the plane to go down. He would write these wonderful narratives from beginning to end with full authority. He would throw out half of the journalism done from the first three weeks of when the plane went down.

DL: Those are the stories people remember.

Engelberg: My wife and I were just talking about a story I know written on [Steiger’s] watch, and it was one of my all-time favorites. It was very early in the cellphone days. It was a snowstorm in Detroit, and the plane landed short of the tarmac and sat for [seven] hours. A passenger pulls out a cellphone. The pilot takes it and gets the name of the head of Northwest Airlines and his home number and gets his wife on the phone. They were a hundred feet away, and they couldn’t get to the tarmac. The toilets were overflowing and people were having panic attacks.

Steiger: I think it was Susan [Carey] who did that, but what it required was her building an atlas of who was on the plane because the airline wouldn’t tell you who was on the plane. She was able to re-create step by step what happened.

DL: That’s the kind of writing you’re talking about.

Engelberg: That’s one kind of writing. I think [writing with attitude and personality] is an equal but different skill that may exist in the same person or may not.

It’s actually funny to watch. We employ Jesse Eisinger, who writes a column for The New York Times and he is very funny and chatty. But he can also do the other. He won a Pulitzer, so he’s got both. He’s probably better at the column. The other is something we kind of nudge him into doing. The habits that don’t allow you do this were training things. If you start by covering the school board, you learn to write in a certain way. The absurdity of the school board meeting that you tell at a cocktail party and leave out of the story should be in a blog post. I think you have to train different generations of people to behave differently. It can be done. I think the value added is on both ends of this. The Web has an enormous amount of opinion journalism, some of which is pretty good and that’s the stuff that people don’t write well for the Web: writing with attitude. The other stuff we are trying to do also needs to be well written.

DL: Can you articulate what that means? You’ve got this fabulous story with huge reporting and now you have to figure out how to present it in a compelling way.

Steiger [pointing to Engelberg]: You are going to be reluctant to do this because you are the hero of this. The Magnetar [Capital] experience, this was what Jesse (Eisinger) and Jake Bernstein got the Pulitzer for. It’s one of the most complicated yet crucially important financial stories that I remember in my entire life in this business. I had nothing to do with it but to watch brilliant people work. The first stage was some incredible reporting by Jesse and Jake to find the proof that these guys had built a great scam. The builders of the scam had probably not broken any laws themselves, but the people around them probably had—but the result was a gigantic fraud that made the perpetrators not millions, not tens of millions, but hundreds of millions of dollars and that drove Merrill Lynch essentially out of business. Even worse, it made the financial bubble last longer and get bigger so that when it finally burst, it was even more disastrous. The complexity, at one level, was very simple, the scheme these guys had. At another level, since most human beings don’t think or live in this world, it was excruciatingly difficult to explain. Steve took that story through something like 30 drafts.

Engelberg: I stopped counting, but there were many drafts. More than I care to remember.

Steiger: What he ultimately did, was to have the first 25 paragraphs contain hardly a fact or quote. It was setup, but it was so brilliantly done. It was the only way to make the story comprehensible. It made the story comprehensible enough that it won a Pulitzer. Secondly, it explained it to the SEC [Securities and Exchange Commission], which launched multiple litigations and the fines have now totaled $400 million.

The reason that this is so important, the story to get it required a deep understanding of highly quantified concepts, and it required deep reporting because so many of these securities did not require SEC filings. There was paper, but only a few people had it. You had to search it out, and then you had to build the theory, and then test it against some of the best lawyers and PR people, including a PR person who used to be one of my top reporters at the WSJ. Then, when you had the story, you had to have the persistence and patience and the insistence to make it readable and to draw the reader. One last thing: In the world we are in now, the computer and data tools we have, we talked about journalism being the first draft of history. In the past, what we did was combine anecdotes. What these tools now allow us to do is to create and test hypotheses and then prove them. For example, if you are white, you have a four times better chance of getting pardoned under the same circumstances, same type of crime and sentence, than a person of color. It’s not from a bunch of examples, but it is a statistically sound database with statistically sound tests run against it. Then what you have to do is go back and get the anecdotes. It’s the anecdotes that help people grasp the story and make people care.

Tofel: [Steiger] is completely right about that. The authority that we now have takes us to a place of writing authority that we aren’t used to having. It’s a generational shift as we take on the power of what we can now do. Big data in journalism allows you to do and say a bunch of things with a level of certainty that you didn’t have before.

Engelberg: It isn’t like there wasn’t data before, but it’s become much more in the fabric of what we do. It’s an expectation of ProPublica that if there’s a major investigation, there will be a data component and we will share it with the readers. Another expectation, whenever possible, that not only will we share overall data, but we will try to share data that is helpful to your actual life. Not only will we tell you about a lot of doctors, but we will tell you about your doctor. That’s a very different way of doing journalism. I think training students to take in both the ability to look for this, to spot it and exploit it, and then write about it in a way the fellow citizens can understand—I think that’s the challenge of journalism education today. We need more of everything: We need more great writers, we need more great number crunchers, and we need more people who can visually display it on the Web in a way that’s compelling.

DL: And that’s not the same thing as someone who can code.

Engelberg: I don’t know if they mentioned the story of [a young employee], but it’s a very interesting story. We hire this woman as a young intern, fresh out of journalism school, and she starts writing and is very good. I think she has real talent. We post that we are looking for someone in the field of data visualization who knows how to code to come in and begin writing programs and doing the work and art and moving into a database. She comes into my office and says that she had a double major in English and graphic arts and had a portfolio, and she asked if she could apply for the job. I asked if she could code and she said not really. I said no, but I wanted to look at her work. Our data guru, Scott, comes to me and says that he’s seen every applicant and everyone can code better than she can, but she’s got the best eye by a mile, and I think we should take a chance on this, hire her and teach her code. At the end of this, we are going to have a journalist who can write her own stories, do the graphics and online and write the code that’s behind it. And now we do. Again, I’m not saying that journalism schools should train people to do this, but if the opportunity arises hopefully you can grasp it. We’ve fended off one large offer so far, and others will come. I’m not saying that is what you need to get from journalism schools. I think there are always going to be a place for people who can write well, and the more you can bring these things together, the better.

DL: Faculty often say that nobody can do everything. How do you address that? Is it specialization?

Engelberg: No one can do all of them. The story about our employee, which is very seductive, is not a model. I remember sitting next to a guy that we have working for us who is almost 70. Anyway, I joined the New York Times as a young person, and he was already old in my eyes. I said Jeff, you and I will always have a job. He asked why, and I said it was because we can find shit out. If we can find shit out, you will always have a job. I think there are parallels to that. Someone who can find shit out and is trained to do that and is good at it will always have a job. Someone who can write a brilliant narrative and take the stuff that has been found out and present it in ways that are believable is going to have a job. People who can navigate the boundaries between the numbers guys and the words guys and understand both and write in a way that’s sensible and clear will have a job. It’s always been this way. Not every journalist came out of a journalism program simultaneously a possible New Yorker writer, a possible investigative reporter, a possible foreign correspondent and a possible science writer. It’s not true.

Tofel: I would say one exception. I actually think, to some extent, that “all” is probably the right answer at a minimal level. For the same reason a good university won’t let you graduate without some ability to write, some mathematical skill, some scientific background, and some foreign language experience. How many graduates are fluent? How many are in advanced science? How many can write for publication? Very few. But how many are supposed to have all of those basics? Everybody. I am on a board that Clay Shirky started the other day of four or five people. The question he posed was, “What about the average person, in their 40s or 50s, in a metro newspaper who is about to lose their job; what would you recommend for them?” It’s a tough question. They should learn some basic coding; they aren’t going to make a team, but their ability to retrain with new skills that are transferable.

Steiger: The coal miner analogy is fair. When they studied coal mining, it turned out it was working in teams, it was manual labor. It turned out to not be that easy to retrain the miners. The guys at metro news have some skills: they can write, they can think, they are fast on their feet and are used to working in groups. There are jobs for them in the world, but it may not be in journalism.

Tofel: The other thing to remember is you need to stay ahead of your peers. Average people’s objective is to stay ahead of most of the people. I think specialization from day one is a bad thing. If you take the postsecondary education analogy, I do think AP [Advanced Placement] is a smart concept. You can let people can skip directly to specialization if they can prove they have the basics in general education.

Engelberg: Let me add a caveat to that. I do have a question about the notion of beginning coding courses. If I were designing a program from scratch, I would say as much as math and physics don’t apply to poets, why don’t we talk about what physics means to us? Beginning coding courses, to me, should begin with how the Internet works, this is the backbone, what front-end and back-end codes are, why things are the way they are. Here’s a coding nightmare, which a small change works out.

Tofel: The best general education courses are like that. Just like they got themselves in trouble with specialization in magazine writing, that’s trouble.

Engelberg: Going back to the point, we had this conversation about buying small, bad video cameras for a reporter, and we wouldn’t do that because in the end, crap videos are crap videos. There are people who really have an eye for it, and they should be the ones that do it.

DL: Will the model of a teaching hospital work in journalism education?

Engelberg: In Oregon, it rains for nine months out of every 12, so the entire staff takes vacation over two months. Our interns are slaves. One guy, whom we hired here, wrote like 50 stories in one summer. I think it’s not entirely impossible to imagine the teaching hospital model working. Newsrooms are desperate for fresh bodies. I don’t think it has to be trivial stuff. I think there are possibilities there.

DL: Is it different from an internship that we already have?

Engelberg: It could be because it could be at a more advanced phase. By the time you get to NYT, ProPublica or the [Washington] Post, you know what you are doing. When you are down at the bottom, straight out of journalism school, you don’t know what you are doing. In the old days, you were able to go somewhere to get screamed at until you knew better.

Steiger: You also hadn’t gone to journalism school. The question is what is journalism school necessary for now?

Engelberg: That is true, but I did have internships which were helpful for learning the business. The problems are down here. The people up here have business model problems, but they don’t have skill problems. The NYT and the Post, their quality stories are as good as ever. The people up here are great. The problem is getting them along this continuum. We hired a young woman who had interned at Mother Jones who is now a full-time reporter for us. When I asked her why she wanted to come here, she said she could either go to a very mediocre mid-size newspaper where no one would pay any attention to me or I could come here. We do have people who have had career paths that come here and then go on to the NYT.

DL: Do you think they aspire to that?

Engelberg: Sometimes. But sometimes they aspire to come here.

Tofel: I would also say that our experience working with young people is that there are some who want to be rich. If getting rich is an important value for you, this isn’t the career. The number of the people who want to be in journalism and want to be rich is a pretty limited set.

DL: What about the programs around the country in entrepreneurial journalism? Is that something you think students should learn? How to pitch?

Engelberg: Let me put it a different way, same thing as the data stuff. I think that, in the larger society sense, it’s useful to know where babies come from. When I was a journalist, I believed that the money fell from a tree onto something called the business side. I couldn’t tell you any fact about our business. None of the facts that became screamingly important to us were in my head. I think to understand the basics of how the business works, where the money comes from is important. For example, I’m on a daily email from Digiday. Today they had a little thing about click-throughs being a lousy way to judge the success of your digital playing field: Here are the reasons, both of them are inadequate. You should have enough knowledge, as a young journalist, to read that article and understand what it’s saying. I think it’s a useful thing so you understand how people are behaving on the Web. Where might the money come from? I think it would be really neat if someone knew enough in a newsroom to say we could do xx, or have you thought of xx, but the notion of pitching an app? I don’t know about that.

Steiger: I think this is question of human nature. I believe, based on my life experience, that most people do not have the entrepreneurial gene.

Engelberg: To open a shop and say, “We will teach you how to be an entrepreneur …”

Steiger: … is like opening a shop and saying, “We will teach you how to be a major league shortstop.”

Steiger: Or you can take someone who has that gene, and give them some information that may be helpful.

Tofel: You can, and probably should, teach someone how to tell a great major league shortstop from someone who will wash out at Class B.

DL: Some people argue that students need to understand and have responsibility for audience engagement. The metric for success isn’t just that it’s a great story or even that it won a Pulitzer. The metric should be this was successful because people keep coming back to it.

Tofel: This is part of Steve’s larger point. It would be critically important for people to understand how their business really works. Today, what is significant about how it works is that the circulation department has been folded into the news department. The skills you used to need only if you worked in the circulation department, a certain number of them you need to have to be successful in the news department, so you should teach those.

Engelberg: And to go back to the shortstop analogy, I think it’s good to be trained and to understand on a visceral level the analytics, so we can measure what happens. We need to know what we are measuring and what success looks like, and those are important questions which people can be taught about. I am imagining a professor standing up and showing different kinds of stories and different writing techniques and how some things succeeded and others didn’t, and why certain presentations work and others don’t. Give people useful tools. On some levels we are all circulation agents now. Then give people the tools to understand when they’re getting it, and why they’re not.

Steiger: One of the things you mentioned a few minutes ago that I want to come back to, because it’s not as good as wheels on suitcases but close. This notion that at the top private schools, and some top public schools, anyone who goes into a Ph.D. track in arts and sciences gets a deal [a fellowship]. If you go into a professional school, whether you are getting a nursing degree or an MBA or a journalism degree, you don’t get anything. There are a few meager scholarships but otherwise you get nothing. If you get accepted into a Ph.D. program in Slovakian languages, you get a full, free ride. Why? Because at universities, faculty have the power. And arts and sciences faculty that have the most power, and they want grad student slaves.

Tofel: And because there’s no money in a lot of those fields. If you didn’t do that, no one would ever study it. One of the lines of inquiry this begs is, have things become so bad yet that journalism people should stand up and say they have become Portuguese, and to have future scholars in Portuguese, you will have to underwrite them.