It is easy to lose sight of why we do things…

I like to write and publish research papers. I think it can be tremendously useful. A research paper serves as a reference describing what was done and what was achieved. Others can learn from these records, verify them, and build upon it.

It turns out that the form is less important than the content. Research papers do not have to be published in established scientific journals. A well-written blog post that describes and idea and reports fully on the results can be worth many formal research papers.

But formal journal publications are also useful. The process of undergoing formal peer review can sometimes enhance the work. Journals also provide some value in increasing the likelihood that the work will remain available for many years.

When I started reading research papers, you had to go (physically) to a library, find the journal you were looking for in the stacks. Figure out where the journal issue you cared for was… then look for the pages where your article was. It was very time-consuming.

Then the web came along and we no longer needed to do that. Search the title of an article and you will find it. To make sure you have the correct version, you can check against the journal’s name, the authors’ names and the publication year. Then if you must have a unique identifier, you can use the article’s Digital Object Identifier (DOI).

Notice how you do not need, ever, the page numbers where the journal article is located. You do not need that information because you are never going to hold in your hand the journal issue on paper. And even if you did, some journals are still printed on paper, after all, there is a nice table at the beginning or end of the issue that tells you where to find the article. Moreover, many journals do not even use page numbers anymore. Once you stop printing the articles on paper, it is a very natural thing to do.

So what you do not need, ever, is the pages where the article is located. You also do not need the name of the publisher or where it is located. Both can be deduced by automated software using the title of the article, the journal’s name and the year of publication, or from the DOI alone. There is no need to add more redundant information that nobody uses.

Five years ago or so, I decided to at least drop page numbers from my list of publications. Ever since it has been a fight with colleagues and bureaucrats who routinely consider my c.v. or my research reports as incomplete because I omit page numbers.

I find that many people don’t even understand what I mean when I object that it is not needed. They have this definition of a complete reference. A complete reference must include page numbers. What is it for? They don’t think about that. They follow (blindly) the rules. What is more, they can’t seem to think for themselves about why they follow the rule.

If I push, people object that there are still edge cases where having the page numbers stated helps. Sure. And there are still cases where horses and carriages are appropriate too.

Sadly, some of these people are established scientists. Believe it or not, some of these people have difficulty putting into question their assumptions. Scientists can be extremely conservative, and not always in the good sense of the word.

It gets worse. Why do I even have to maintain a publication list? It is fairly easy to find out what I published without me having to tell anyone. Google maintains automatically a list of my publications.

These are simple and small things, of course. But it does not stop there. We are also stuck in our ways with more important issues.

For years, I have been advocating what I call Star Trek economics. The idea is simple. If you go back a couple of centuries, everyone was a farmer or a servant working for a lord. Then the industrial revolution came along and people got jobs. Then things got a lot better over time because workers got more and more productive. Education, training, better tools… everything conspired to make workers ever more productive.

Today, we have solid evidence that this has come to an end. The same way the lords and their peasants could not go on in light of the industrial revolution, we also cannot go on “holding on jobs” the way we used to.

We need to rethink our economy. Some people have been proposing the adoption of basic income: everyone gets a paycheck from the government. When trying to promote this idea, I have gotten a tremendous amount of resistance. People are generally fine with some notion of “welfare” (if you can’t work and you have children to feed, we will help you). But thinking out of the box is painful.

Why can’t we go on listing page numbers for research articles, and training for good old jobs where we shall be ever more productive? You can go on. Forever if you want to… but you will increasingly look like an automaton that runs over the same algorithm, again and again, without any critical thought.

Truth is, it takes a lot of hard work to keep track of the reasons why we do things. And everyone is lazy, smart people especially.

Published by

Daniel Lemire

A computer science professor at the University of Quebec (TELUQ).

13 thoughts on “It is easy to lose sight of why we do things…”

  1. What you are describing is strongly related to status quo bias. In many cases, people agree that there is something wrong with the way things are being done. However, when you suggest a radically new way of doing things, no matter how sensible it is, many people just instinctively reject it. Often the reason they give is something vague like “I just don’t think it would work”. It is very sad that good ideas never get implemented just because they are too much out of the ordinary.

    I think that to cure status quo bias, people need to educate themselves about it. I can recommend Daniel Kahneman’s book “Thinking fast and slow”. After studying the phenomenon myself, I can now more easily recognize a situation where I’m about to reject something just because it’s too much out “out-of-the-box”.

  2. I suspect there needs to be a balance between conservatives and risk takers. We obviously want free thinkers to take chances and try new things so we can enjoy the fruits of innovation, but we need conservatives to test these ideas against the status quo, critically examining them to see if they provide significant benefit – worth abandoning what we hold as “the way” at this point in time.
    It is often hard to see the other’s point of view, and even harder to let go of one’s own in the process, but demonizing the other does not help.

  3. > Some people have been proposing the adoption of basic income: everyone gets a paycheck from the government.

    I’m pretty open to this idea but it looks like the meaning of “basic income” differs depending on which proponent of BI you talk to.

    Some BI proponents seem to be fine with means-tested BI (e.g. negative income tax) but others seem to oppose the idea of means testing.

    And even the BI proponents of the latter camp say “even free-market economists like Milton Freedman supported BI” but as far as I know he (or other economists) never supported “even billionaires get it”-form of basic income. He merely said negative income tax could be more effective than the current ragbag of welfare programs.

    Much of human labor is automated (which I think is a good thing) but we are nowhere near the point where human labor is obsolete (robots aren’t stocking shelves at supermarkets or collecting garbage) so some say Basic Job Guarantee (the government gives a basic minimum wage job for anyone who asks for one) is a better form of welfare before we reach that point.

    1. The difference between a negative income tax and an “everyone gets it” basic income is all about semantics. In either case, people with high income pay more than they receive, while people with low income get more than they pay.

      Basic Job Guarantee (the government gives a basic minimum wage job for anyone who asks for one) is a better form of welfare before we reach that point.

      We had such a thing, but we got rid of it…

  4. I also agree that you shouldn’t have page numbers on CVs/references, but on CVs, the number of pages is often still important to show the “heft” of the contribution. This can easily be calculated from the PDF, thankfully.

  5. @Daniel
    > workhouse
    Yeah, or the FDR plan. We don’t live in a world where all the public roads/parks/government buildings are impeccably clean so creating jobs (that even the most advanced robots by Boston Dynamics still can’t do even remotely well) would be trivial for a while.

    Not that I advocate BJG (I advocate experimentation of different programs to see which one seems to work best) but I don’t think Basic Income vs Basic Job debate is hardly resolved:

    > The difference between a negative income tax and an “everyone gets it” basic income is all about semantics

    I think it’s true to some extent, but the problem of getting rid of means testing seems to be that to give any meaning amount of BI to every single citizen, it probably requires significant increase in taxes rich people pay, and that could negatively affect their behavior. NIT doesn’t seem to suffer this problem as much.

    Also, people like Bill Gates or economist Robert Frank support progressive consumption tax (as opposed to income tax), do you have any thoughts on this?

    1. We don’t live in a world where all the public roads/parks/government buildings are impeccably clean so creating jobs (…) would be trivial for a while.

      We have people right now doing these jobs.

      1. We do, but the government can hire more people to make these jobs easier, right?

        And it’s not just janitorial jobs. Lots of e.g. elementary school teachers (or simply people who supervise kids’ assignments at school) can be hired. Unemployed skilled workers can become public auto insurance company workers at minimum wage (in some cases this may save some taxpayer money).

        1. Sure, today a robot cannot clean a park or serve as a substitute teacher. So people who can clean parks or act as substitute teachers can be considered skilled. And we do not have, today, a technological unemployment problem.

          People like me who advocate basic income think about 10 years, or 20 years, out in the future… when we do have cheap robots that clean parks, drive cabs and act as substitute teachers better than human beings can.

          1. Completely agree. Eventually a large part of complex tasks like cleaning roads/buildings/etc or educating kids will be done without much human labor, and when that happens basic income makes a lot of sense. I don’t advocate making up “bullshit jobs” as discussed by David Graeber.

            1. I think it will make sense to create “bullshit jobs”… like have a human park maintenance supervisor even if robots can do that job better… I think human beings will keep on attributing themselves titles even as computers take over… kind of like the lords from two hundred years ago… we will like to be nominally in charge even if we are not needed.

              You will go to the hospital… a computer will diagnose your problem, possibly explain to you how you can get a cure, but you will still, formally, have a doctor “in charge” of your treatment…

              These people will have long meetings to decide what color we should use to paint the shed…

              1. Well, maybe creating “bullshit jobs” isn’t so bad but as long as they create some value to the public they aren’t bullshit jobs in my book.

                > a human park maintenance supervisor

                I don’t know. Don’t we need someone in case e.g. dogs chew on the wiring of janitor-bots?

                I don’t think even future technologists can plan for all the potential externalities that can mess with their software/hardware, just like people who build tall buildings (even today) can’t/don’t plan safety measures in case some attackers fly airplanes into the building.

                I recently listened to an NPR episode about firefighters and their declining value as perceived by the public (residential fires simply don’t happen very often anymore).


                But most people (including myself) probably aren’t very comfortable getting rid of their jobs. In this way, their jobs aren’t bullshit, even though they do nothing most of the time.

                1. But most people (including myself) probably aren’t very comfortable getting rid of their jobs. In this way, their jobs aren’t bullshit, even though they do nothing most of the time.

                  That’s a very good example. Even if we had robot firefighters that could deal with fires better than human beings, we might still want to have human firefighters, if only to supervise the robots.

Leave a Reply

Your email address will not be published. Required fields are marked *

To create code blocks or other preformatted text, indent by four spaces:

    This will be displayed in a monospaced font. The first four 
    spaces will be stripped off, but all other whitespace
    will be preserved.
    Markdown is turned off in code blocks:
     [This is not a link](

To create not a block, but an inline code span, use backticks:

Here is some inline `code`.

For more help see