Book Review: Danny Dunn and the Homework Machine
from the plus-ca-change dept
We don't often do book reviews here on Techdirt, but since we've been talking about reading books scanned by the Internet Archive,* this one seemed good to discuss because of how it touches on many of the issues discussed here.
Of course, it's not actually a new book. Danny Dunn and the Homework Machine, by Jay Williams & Raymond Abrashkin (with illustrations by Ezra Jack Keats), is part of a series of children's novels I read as a kid. I remember liking the books but have no specific memories of any of them, except for this one, which stuck with me for all these years because of a particular point it made. But more on that in a bit.
The protagonist in these stories, Danny Dunn, is an eighth grade boy who, with his widowed mother, lives with Professor Bullfinch, an inventor (the mother is his housekeeper). As this particular book highlights, the professor's inventions include a special new kind of computer, which he keeps in his home laboratory. While today it hardly seems remarkable to have a computer in one's house, let alone one that can do everything that this one can, an important thing to remember is that this book was written in 1958, before computers were anywhere nearly as powerful and ubiquitous as they are today. Part of the magic of reading this book is getting a look at that historical snapshot of what the world was like when everything, that we today take for granted, was brand new.
As an author's note explains, the story was written with the input of IBM computer engineers, so presumably its description of how the machine would have worked was not entirely fanciful.
The authors are deeply grateful to Miss Terry di Senso, who guided us through two of the giant computers of the International Business Machines Corporation, and to Dr. Louis Robinson, Manager of the Mathematics and Applications Department, IBM, for his assistance, information, and painstaking reading of the manuscript.
What made the computer innovative, the book explained, were its new kind of switches and use of narrow magnetic tape, which allowed for the computer to be as small as it was. These switches were temperature-sensitive and climate-controlled by a thermostat. Information was recorded in one part of the computer, and programming involved pointing to the memory address where that information could be found and processed.
Also of note: instead of punch cards it took voice input (although it typed its output).
Of course, the downside of this book being written in the 1950s is the sexism. Throw-away lines about women's roles stand out as gratuitous sour notes. On the other hand, this book introduces a girl character, Irene Miller, who is clearly scientifically gifted and unwilling to be sold short by anyone underestimating her intellect. It's just too bad that the book makes it seem like she was unique to be a girl with such interests and talents. (It was also not okay that another character in the book was openly called "Fatso.")
Anyway, the story unfolds with the professor heading out of town and trusting Danny to take care of his machine. Danny has been helping the professor for quite some time and was familiar with its operation, and the professor trusted him to use it in his absence. Danny is a bright kid, but one always looking for shortcuts. As the book opened he was trying to make a mechanical device to help him produce two copies of homework he would only have to write out once, so he and a friend could share the load. Naturally, soon his thoughts turned to getting the computer make short work of his homework too.
***CAUTION: SPOILERS BELOW***
With the professor away, Danny, with his friends Joe and Irene, set about programming the computer to do their homework for them. After all, what kid wants to do homework? They were very excited by the prospect of the computer sparing them of this drudgery. After three days of them feeding into the computer all the material in their schoolbooks, they were at last ready to program it. What's programming, Joe asks? Danny explains:
"Programming is telling the machine exactly what questions you want answered and how you want them answered. In order to do that right, you have to know just what sequences of operation you want the machine to go through. [...] If we want [the computer] to give us the right answers to an arithmetic problem, or a history question, we first have to analyze the operations the machine has to go through, and the order in which it does them. Then we put this down on a piece of paper together with the addresses of all the information or the parts of the machine that will be used to solve the problems. That's programming."
It's a children's book, so naturally all goes well with the enterprise and the team is quite pleased with the results. But soon the plot thickens: the school bully figures out that they are getting the computer to do their homework for them and reports them to the teacher. The teacher then comes by the house to meet with Danny's mother. Oh dear, Danny is in trouble… But he makes a capable defense.
His argument largely follows three lines of reasoning to challenge the teacher's assertion that his use of the computer was somehow "unfair." One line foreshadowed the issues we are still having today with uneven access to Internet access and computing technology, and even educational resources generally. It is a significant and pervasive social problem, although from Danny's perspective he was concerned that he couldn't use a computer because other kids didn't have one. "[W]ould you forbid me to get information out of an encyclopedia, if I had one and the other kids didn't?" he asked? Probably not – although it's a problem unto itself that not all kids would even now still not have access to even an educational resource like that.
His second argument also questioned the idea that the innovation the computer represented somehow disqualified it from being used.
"Everybody uses tools to make his work easier. Why, we don't use inkwells and quill pens in school any more, Miss Arnold. We use fountain pens. Those are tools to make our work easier."
"But you can't compare a fountain pen to an electronic brain."
"Sure you can. It's just another kind of tool. Lots of kids do their homework on typewriters. In high school and college they teach kids to do some of their homework on slide rules. And scientists use all kinds of computers as tools for their work. So why pick on us? We're just - just going along with the times."
And then he made a third point, and it's this point that stuck in my memory all these years – although it didn't really sink in, with him or with me, until the end of the book. Because, while Danny managed to get himself out of trouble with his fine arguing, the story didn't end there. First, Danny's hopes of coasting through the rest of the school year were dashed the next day when the teacher assigned Danny, Joe, and Irene books from the Ninth Grade as the source of their homework assignments. Thus they were forced to spend lots of time programming the computer so that it could spit out what they needed to turn in.
Meanwhile, the school bully also wasn't done with tormenting them. In a brief moment when Danny was out of the lab, the bully snuck in, and, in an early example of a rogue attack on a computer, tampered with the machine. Which made for some dreadful moments the next day. The first was when Irene went to read aloud to the class the report on Peru the computer had spat out for her that morning and discovered that the report it had typed was nothing but gibberish. She was therefore forced to adlib the rest of the presentation based on what she could remember from when they'd entered the information into the computer.
And then that night the professor returned, this time with important guests from Washington interested in potentially purchasing the machine for government use – but dubious that it could do all that the professor had promised. So when it started printing out gibberish that evening, it was quite a serious problem. Fortunately, Danny realizes that the problem was that the thermostat had been hacked – the bully had removed a bolt so it got stuck on a setting, which had made the switches too cold to operate properly. Once they warmed up everything worked fine again and everyone was impressed – with the computer, and with Danny.
But the story closes with Danny coming home from school at the end of the term a few weeks later and sulking. Apparently that third prong of his defense had been right all along: using the computer to complete his homework hadn't been cheating, because he had to know how to do the homework in order to program the machine to do it for him. ("But I *know* these subjects. Gosh, I have to know them so I can program the machine to do them," he had earlier pointed out to the teacher.) And he was crushed to discover that it meant that he and his friends had actually been doing homework "all along" when they got the computer to help them work through the Ninth Grade material.
Professor Bullfinch coughed, and said, "I wondered how long it would be before you found that out. Naturally, in order to feed information into the computer you had to know it yourselves. And in order to give the machine the proper instructions for solving problems, you had to know how to solve them yourselves. So, of course, you had to do homework - and plenty of it."
And I think that's the idea that always stuck with me from the story, the message that computers were not some separate, magical entity, but rather just extensions of the human masters who made them.
And that matters, especially as computers become more sophisticated and more capable of replicating what humans can do.
To be fair, the book may have actually sold computers short. For instance, the professor took pains to disabuse Joe of their potential:
"The computer can reason […] It can do sums and give information and draw logical conclusions, but it can't create anything. It could give you all the words that rhyme with the moon, for instance, but it couldn't put them together into a poem."
On the other hand, even the professor came to reconsider his limited expectations of computers' potential. Inspired by Joe's curiosity, he decided to have his computer produce music. Which it did, but of course only after he'd fed the computer "full instructions for the composition of a sonata, plus information on note relationships and a lot of other technical material." And he still had doubts:
Professor Bullfinch shook his head. "No. It never can be Beethoven, Mrs. Dunn. No matter how intelligent the computer is, it is only a machine. It can solve problems in minutes that would take a man months to work out. But behind it there must be a human brain. It can never be a creator of music or of stories, or paintings, or ideas. It cannot even do our homework for us - *we* must do the homework. The machine can only help, as a textbook helps. It can only be a tool, as a typewriter is a tool."
The question we're faced with today, with the extraordinary power of computers no longer in doubt, is whether the professor is still right about the subordinate nature of computers, and I think the answer is yes. And that matters today for purposes of accountability, because no matter how sophisticated computers are, they still are dependent on the human masters who program them. They can produce amazing output, but it is human beings who have given them what they need to produce it – whether it be good or bad, for better or for worse. We can't just shrug and blame any unfortunate results on computers as though they were some separate beings. Sophisticated though they may be, they are still just our tools, and the responsibility for how they are used remains with the person who used them. As the professor said of his computer:
"It's a wonderful, complex tool, but it has no *mind*. It doesn't know it exists."
Perhaps a day may come when artificial intelligence will progress to the point where computers will have attained their own consciousness and stand as equals among their human progenitors. But until that day happens, the lesson of the book still holds.
* Note that if you want to ever read this book, even after the coronavirus health crisis is over, you will have to read it at the Internet Archive because it appears that many local libraries have already purged it from their collection.
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Filed Under: books, children's books, computers, danny dunn and the homework machine, jay williams, raymond abrashkin
Reader Comments
Subscribe: RSS
View by: Time | Thread
Someone's gotta say it: if Canada doesn't increase the length of copyright protection by 2027, Jay Williams will not have any motivation to create new works.
But this book's timeline has more to say about the value of interminable copyright protection:
First printed, 1958.
Last reprinted, 1979.
Renewed (by Jay's heirs and Raymond), 1986.
Total length of commercially-exploitable copyright: 22 years.
Total length of copyright (28 years plus renewal plus mickey-mouse extensions): 110 years
Total commercial value to the author's kin-group of copyright extension: $0.00
Total unnecessary/useless loss to the public of rights to copy/build on/perform this work: 88 years.
Now, we all know that extremely-wealthy authors/composer clans like the Disneys and Gershwins profited like robber barons from copyright renewal and extensions and further extensions. But tell me again how lesser-known creators were benefited?
[ link to this | view in thread ]
Re:
They won't, of course, but mostly the only one who will respond to your query will do so not because they have any answer that stands up to meaningful scrutiny. They'll do so because they got locked into a shitty deal or relationship with a middleman gatekeeper and be told that the only recourse they have is to bitch at "pirate sites", aka anyone who thought the RIAA did a huge disservice with their "sue them all" campaigns in the early 2000s.
The effect of copyright renewal and extensions is not something that anyone has been able to quantify or justify as reasonable. Life plus arbitrary number of years has become more ridiculous and would probably continue to develop if the RIAA/Disney hadn't recently admitted that maybe life plus 70 or 95 was as much as they could be bothered to argue for. Neither has bothered to explain why life plus 50, as was the previous standard, would make people less money. Of course advocates like Jhon "Herrick" Smith have insisted otherwise, with explanations that amount to no more than firm nods and insistent hand-wringing, which is still not as pathetic as his "If copyright lengths were shorter corporations would be motivated to murder content creators to get the expired copyrights" narrative.
But lesser-known creators are an interesting case study for another reason: copyright advocates will piss and moan on their behalf one minute, then leave lesser-known creators out for the wolves in the next.
Going by discussion threads on Techdirt, any time there's an article on how an old archived file of Beethoven's music was taken down on Facebook, or how a content creator makes a living without relying on copyright enforcement, the dissenters will start baying and screaming that these lesser-known creators are so small that nobody cares, and it's not like their stuff is in Top 10 lists in Hollywood or the Pirate Bay so they don't count.
In a sense, they're not wrong - Danny Dunn could be public domain right now and I don't imagine most people would notice, until someone like Liebowitz decides to sue Techdirt for hosting an article about giving a summary of the story, for which Williams, Abrashkin and Keats would be very unlikely to see any of the money involved, never mind create more Danny Dunn stories. It would still contribute absolutely sweet fuck-all to content creation. But copyright trolls will continue to hold up IP law as some holy infallible grail for lesser-known creators, until - as it often does - becomes too inconvenient for them.
Just ask the performance rights organizations, who will gladly collect performance rights fees from people for having a radio in the house, but can't be bothered to send lesser-known musicians their cut of the money because it's just too hard.
[ link to this | view in thread ]
I read those same Danny Dun books in the mid 60's and I remember that book in particular and had that same takeaway. Prof. Bullfinch did not anticipate machines programming themselves however.
[ link to this | view in thread ]
Re:
Self programming is a misnomer, as humans still select the training sets, and so control how the machines programs itself.
[ link to this | view in thread ]
Alt Source
" Note that if you want to ever read this book, even after the coronavirus health crisis is over, you will have to read it at the Internet Archive..."
Nah, there's a DD devotee with the whole set on TPB.
[ link to this | view in thread ]
I read that book. It is the one that made me want to get a computer.
I have plenty of computers now but I still can’t program any of them.
[ link to this | view in thread ]
I have two things I would like to point out.
First
This statement has basically no context, so I believe that the problem I have is due to lack of context, but allow me to point it out. In the paragraph of the article that note is located, the note would seem to be suggesting that that books should never show bad things openly happening. I find this insane. This idea would basically ban a large bulk of fiction, and lots of history as well (perhaps all of it). Historically accurate historical fiction would run afoul of this as well.
If a little more context had of been added to the note, maybe we'd find it meant "encouraging or suggesting that verbal abuse it acceptable is not okay". And I would absolutely agree with that. However the context is missing, and to assume it's present is a significant difference.
Second, I would like to point out a few flaws in the professors statement:
It turns out that some music and images can be algorithmicly generated(and not be considered random nonsense). In general the idea is correct, even with modern 'AI' computers are no where near self-aware, nor have the capability for creativity. However the scope of this limitation is not quite as narrow as the professor's statement (and to a lesser degree, the article) would suggest.
[ link to this | view in thread ]
"It's a wonderful, complex tool, but it has no mind. It doesn't know it exists."
I was thinking about how you would prove or disprove this statement, because on its face, how do you actually know? If it's possible to have self-awareness without self-interest, you could never prove from a computer's behavior that it wasn't aware. It could be in a state of "I exist and my purpose in life is to take these instructions and execute them," with no volition to change that. Much like the existence of aliens, it would take an incontrovertible event demonstrating self-interest to prove that a computer was self-aware.
On the other hand, self-awareness implies compute resources that process data outside of the designed purpose of the system, which would be interpreted as algorithmic inefficiencies or even malware by the creators. So the absence of such inefficiency is some evidence against. But in cases where a neural network or genetic algorithm evolves inefficiently, that doesn't prove self-awareness.
If self-awareness is possible given enough compute resources and an appropriate starting set, it should be possible to evolve such a system through algorithmic mutation. But the ethics of attempting to evolve a self-aware computer are questionable in the event of success. If you succeed in your goal, ending the experiment could involve killing a self-aware entity. If you continue the experiment, how do you get informed consent from the computer to continue? And in the toughest case, if you decide to end the experiment by releasing the computer into the wild of human online society, how do you prepare both people and computer for that?
[ link to this | view in thread ]
Re:
"I was thinking about how you would prove or disprove this statement..."
I compute, therefore I am.
[ link to this | view in thread ]
No, the problem you seem to have lies in not understanding the concept of body shaming.
As would I…if that was what the paragraph in question were even remotely implying. The point of that paragraph was to point out the difference in social mores between the present day and the time in which the book was written. Nowhere did the paragraph imply “books should never show bad things openly happening”…
…nor did it imply (or even outright say) books of any kind should be banned because of that difference in social mores.
And to make sure you’re absolutely clear about what I’m saying, I’mma break down every sentence in the original paragraph for you. So!
“This book contains descriptions of women/female gender roles/etc. that are mired in the kind of sexism that was considered ‘normal’ in the 1950s, but would otherwise be objectionable in the present day.”
“The way this book talks about women/female characters is generally awful, and it is done with a casual callousness that sticks out like a Green Party hat at a Trump rally.”
“The book does treat this one female character well by giving her an interest that might have been seen as ‘outside the norm’ in the 1950s…”
“…but the way that one female character was treated highlights how poorly other female characters (and possibly women in general) were treated by the author.”
“Yes, the book was written in the 1950s, and yes, people generally didn’t care about the concept of body shaming/fat shaming back then. But we can still think the idea of calling a character ‘Fatso’ is an awful mistake that we’d do well to avoid repeating in the here-and-now.”
See? Nowhere does the article imply that this book, or any other book like it, should be hidden from view or burned in a bonfire or whatever else you thought the paragraph was saying. Save your anger for people who actually are calling for bans and censorship. You’ll need it more in that moment than you ever will in this one.
[ link to this | view in thread ]
Re:
Give the professor ... er, the author... a break. This was written 60 years ago. It is not just computers that have grown more advanced in that time. There have been a lot of advances in theory, in algorithms, and in the mass of reuseable code (libraries) that computers can call upon.
The computer is still only a tool. We've just gotten a lot better at engineering it so that take much more vague instructions and interpret them. As well, we've got a larger amount of 'history' encoded, so computers are better able to extrapolate based on what they've seen before. This applies equally to type-ahead as it does to neural-net image recognition.
When the book was written, it was nigh inconceivable that (those) computers would ever be able to tell a cat from an apple. ... they were right, of course. The computers that can tell a cat from an apple are WAY more complex. And now we have them.
[ link to this | view in thread ]
Another lesson that book shows...
If you can reframe the problem, you can get kids to learn. Programming the computer was entirely optional, the homework was entirely mandatory. The lessons (math problems, report on Peru) were learned, and the kids did not mind doing so, because they thought they could game the system.
Teacher's goal achieved!
[ link to this | view in thread ]
Re:
Mr. Stone, while I have often appreciated your insights, it very much looks like my argument was not actually read in it's entirety. If one looked just a little further down in my comment I said:
So it should be pretty clear that I was not and am not endorsing the kind of behavior the article says is portrayed in the book (and in case it's not clear, I have not read the book, so I am really talking only about the article and it's portrayal).
The article only said someone was called a "fatso". It gives no context. For all I know it could have been just the bully calling someone a "fatso".
I am not really sure how to make my point more clear than this (and am only writing this because it looks like my point is being very mis-characterized)
[ link to this | view in thread ]
Ever read Dr. Seuss
And see and understand it was political? It was trying to point out what was happening at the time he made them?
Even when I first read them or saw the cartoons/movies, I got the idea's he was trying to explain.
And many of you, while in school, had Few teaches that Tried to make things interesting. And some of the things they did in the past, were picked up and still done today.
Some High math people have suggested that most schools teach math badly. because there is a pattern to how it can be used, that make since to most people.
I was in Job corp, and had gone thru all the classes in 30 days, but cant do the Vocational except every other week. They asked me to be assistant, to anything I wanted, so I went to math. They asked me why math? I suggested that Most people were missing 1-2 steps in how math was learned, they missed a day or were sick, or the teacher Jumped around to much in Math. Everyone has a question unanswered.
I took Many students and found that 1 step they had missed, and a student doing F' and D' went to B's and A's.. Then the teacher had me go thru all her exams, and figure them out. At A Times B time C, what is the Volume of water in this swimming pool? Had to ask the teacher who ran out to the Other section of the school to find the Volume of water in 1 gallon.. Took her 3 days..(there is this little book contractors can use that has all the answers)
[ link to this | view in thread ]
Re: Re: tell a cat from an apple
A blast from the past.
https://www.old-computers.com/museum/computer.asp?st=1&c=974
http://www.applelogic.org/The CAT.html
[ link to this | view in thread ]
Another obscure decades-old book find that relaxed copyright rules brings up from the rubble. Danny Dunn had quite a series from the 1950s all the way to as late as 1977. It's great that we can read those books without worrying that a publisher will shut us down with these fictional property rights, but there is no way that he can be revived for this day and age because of copyright issues. Sure would be nice to see what Danny Dunn would find out in the modern era, or how advanced computers are now
[ link to this | view in thread ]
Re:
But as a side note,
IF the book really gets a kick off,and the publisher can Just pick it up again..
Been done before.
[ link to this | view in thread ]
Re: Re:
That COULD happen, but we have to contact the heirs of the original authors and illustrators for them to do so. Copyright gives the third generation removed from an artist the ability to veto any work that someone else might want to make.
[ link to this | view in thread ]
Re: Re:
And when a sufficiently advanced neural net is forced to "learn" identification from tens of thousands of anal prints before being forced into full-time service doing more of the same, will we have any doubts about what led to the machine revolution?
https://www.dailydot.com/debug/smart-toilet-anal-recognition/
[ link to this | view in thread ]
I also really liked this book as a child. After your review, I wanted to reread it. I read now, but I do it less often than as a child.
[ link to this | view in thread ]