I have preordered, and am looking forward to reading my copy when it arrives. Seems like a way to buy lightcone-control-in-expectation-points very cheaply.
I admit I’m worried about the cover design. It looks a bit… slap-dash; at first I thought the book was self-published. I’m not sure how much control you and Eliezer have over this, but I think improving it would go a long way toward convincing people to spread it & its ideas as mainstream, reasonable, inside-the-overton-window.
+1 on the cover looking outright terrible. To make this feedback more specific and actionable:
If you care about the book bestseller lists, why doesn’t this book cover look like previous bestsellers? To get a sense of how those look like, here is an “interactive map of over 5,000 book covers” from the NYT “Best Selling” and “Also Selling” lists between 2008 and 2019.
In particular, making all words the same font size seems very bad, and making title and author names the same size and color is a baffling choice.
Why is the subtitle in the same font size as the title?
And why are your author names so large, anyway? Is this book called “If Anyone Build It, Everyone Dies”, or is it called “Eliezer Yudkowsky & Nate Soares”?
Plus someone with a 17-character name like “Eliezer Yudkowsky” simply can’t have such a large author font. You’re spending three lines of text on the author names!
Plus I would understand making the author names so large if you had a humungous pre-existing readership (when you’re Stephen King or J. K. Rowling, the title of your book is irrelevant). But even Yudkowsky doesn’t have that, and Nate certainly doesn’t. So why not make the author names smaller, and let the title speak for itself?
I understand the artistic desire to have the irrecoverable red event horizon of superintelligence to underscore the “would kill us all” part, but since it makes the words “kill us all” harder to read, I’m not sure if this current design underscores or obscures that.
Surely it would’ve been possible to somehow make title and subtitle less than four lines of text each?
And overall, the cover just looks very cheap and low-effort.
EDIT: More fundamentally: in all media, title & cover art are almost as important as content, because you can’t get people to click on your video, or to pick up your book in a bookstore, if title & cover aren’t good. Then it doesn’t matter how great the content is, if nobody ever sees it.
Anyway, the title is good, the cover is bad, and I can’t assess the content yet. You say this book has been in the works for over a year, and that you spent lots of effort on polishing its content. Then if it’s a good fraction of MIRI’s output for that time, and e.g. a cover is responsible for (say) 20% of a book’s impact, then wouldn’t that justify spending >>>$100k on the cover design? This one looks more like a cover purchased on Fiverr.
Also see this discussion on the need to spend a significant fraction of one’s effort on a piece of content, on stuff like title and cover and thumbnail and book blurb.
The “lightcone-eating” effect on the website is quite cool. The immediate obvious idea is to have that as a background and write the title inside the black area.
If one wanted to be cute you could even make the expansion vaguely skull-shaped; perhaps like so?
If you care about the book bestseller lists, why doesn’t this book cover look like previous bestsellers? To get a sense of how those look like, here is an “interactive map of over 5,000 book covers” from the NYT “Best Selling” and “Also Selling” lists between 2008 and 2019.
Most of those are fiction or biographies/memoirs (which often have a picture of the subject/author on the cover), which seem to have a different cover style than other books. Skimming through some lists of NYT bestsellers, some books with the most comparable “Really Big Thing!” topics are “Fascism: A Warning” (Madeleine Albright, cover has large red-on-black lettering, no imagery), “How to Avoid a Climate Disaster” (Bill Gates, cover has large gradiented blue-to-red text on white background, author above, subtitle below, no imagery), “Germs” (title in centered large black lettering, subtitle “Biological Weapons and America’s Secret War” in smaller text above, authors beneath; background is a white surface with a diagonally-oriented glass slide on it), and “A Warning—Anonymous” (plain black text on white background, subtitle “A Senior Trump Administration Official” in small red lettering below, no imagery). Neither cover version of IABIED looks that different from that pattern, I think.
Given that the book is being published by a major publisher, it can safely be assumed that the cover design was made by a professional cover designer, who knew what they were doing.
Contrary to what you wrote, the title has a bigger font size than both the subtitle and the authors’ names (this is true of both the American and UK covers; I am primarily talking about the American cover, which I presume is the one you are referencing). Even if the author names were the same size as the title, it is immediately obvious which one is the title, and which one isn’t. Putting the subtitle in a dark grey, which is much closer to the background color (black) than the color of the title (white) is, also does a lot to move emphasis towards the title of the book (away from the subtitle)
Most importantly, the title is plenty big. If it was small, then I would feel there is something to what you are saying; but the title is quite large and readable from a distance, and clearly delineated from the rest of the text on the cover.
In this case, part of the point of publishing a book (including writing it in the first place), is presumably to promote the identity of the authors, to make them a known name / schelling point for discussion about AI safety. That would indicate making the names quite prominent on the cover.
I see that the numbers indicate people disagree with this post. Since there are several clauses, it’s hard to know which specifically (or all of them) are being disagreed with.
The second paragraph (beginning “Contrary to what you wrote...”) is a list of factual statements, which as far as I can tell are all correct.
The third paragraph (“Most importantly, the title is plenty big...”) is more subjective, but I’m currently not imagining that anyone is disagreeing with that paragraph (that is, that anyone thinks “actually, the title is too small”).
The fourth paragraph (“In this case, part of the point...”) is more speculative, and I could easily imagine someone reading it and thinking “that’s not the point of publishing / writing a book”. There’s certainly a reason I put a “presumably” in there. I do still feel that there’s something to what I’m saying in that paragraph. My surprise would be of a limited extent if Soares and Yudkowsky said “that was not a consideration in our decision to do this”—but I would be somewhat surprised.
I can see someone disagreeing with the first paragraph (“Given that the book...”), but my current state of mind is that such people would be simply wrong. The book is not being self-published, but is being published by Little, Brown and Company. Some excerpts from Wikipedia’s article on Little, Brown and Company:
“The company was the most extensive law publisher in the United States, and also the largest importer of standard English law and miscellaneous works, introducing American buyers to the Encyclopædia Britannica, the dictionaries of William Smith, and many other standard works. In the early years [starting in 1837] Little and Brown published the Works of Daniel Webster, [...], [and] Letters of John Adams. [...] Little, Brown and Company was the American publisher for Edward Gibbon’s The Decline and Fall of the Roman Empire.”
The firm was the original publisher of United States Statutes at Large beginning in 1845, under authority granted by a joint resolution of Congress.
and
During this time [about 1908 to 1985] the joint Atlantic Monthly Press/Little Brown imprint published All Quiet on the Western Front, Herge’s The Adventures of Tintin, James Truslow Adams’s The Adams Family, [...] [and] J. D. Salinger’s The Catcher in the Rye.
The point being, the company that is publishing Soares and Yudkowsky’s book, is an established company that has sold important and/or bestselling works for two centuries. The people there know what they are doing, and that includes the people who design covers, as well as the bosses of the people who design the covers.
I imagine most disagreement comes from the first paragraph.
The problem with assuming that since the publisher is famous their design is necessarily good is that even huge companies make much worse baffling design decisions all the time, and in this case one can directly see the design and know that it’s not great – the weak outside-view evidence that prestigious companies usually do good work doesn’t move this very much.
Yes, my disagreement was mostly with the first paragraph, which read to me like “who are you going to believe, the expert or your own lying eyes”. I’m not an expert, but I do have a sense of aesthetics, that sense of aesthetics says the cover looks bad, and many others agree. I don’t care if the cover was designed by a professional; to shift my opinion as a layperson, I would need evidence that the cover is well-received by many more people than dislike it, plus A/B tests of alternative covers that show it can’t be easily improved upon.
That said, I also disagreed somewhat with the fourth paragraph, because when it comes to AI Safety, MIRI really needs no introduction or promotion of their authors. They’re well-known, the labs just ignore their claim that “if anyone builds it, everyone dies”.
These aren’t really finished quality either, but the authors should feel free to borrow and expand on any ideas they like if they decide to do a redesign.
I only like the first one more than the current cover, and I think then not by all that much. I do think this is the sort of thing that’s relatively easy to focus group / get data on, and the right strategy is probably something that appeals to airport book buyers instead of LessWrongers.
Finally created a LW account (after years of lurking) to upvote and agree on the cover design issue.
This is a topic I read whatever I can get my hands on, and if I saw this book in a store (and did not know EY or Nate), even I would be a bit put off to give it a read
Given the stated goal of trying to make this a bestseller, I feel like the cover is a pretty big impediment
I have preordered, and am looking forward to reading my copy when it arrives. Seems like a way to buy lightcone-control-in-expectation-points very cheaply.
I admit I’m worried about the cover design. It looks a bit… slap-dash; at first I thought the book was self-published. I’m not sure how much control you and Eliezer have over this, but I think improving it would go a long way toward convincing people to spread it & its ideas as mainstream, reasonable, inside-the-overton-window.
+1 on the cover looking outright terrible. To make this feedback more specific and actionable:
If you care about the book bestseller lists, why doesn’t this book cover look like previous bestsellers? To get a sense of how those look like, here is an “interactive map of over 5,000 book covers” from the NYT “Best Selling” and “Also Selling” lists between 2008 and 2019.
In particular, making all words the same font size seems very bad, and making title and author names the same size and color is a baffling choice.
Why is the subtitle in the same font size as the title?
And why are your author names so large, anyway? Is this book called “If Anyone Build It, Everyone Dies”, or is it called “Eliezer Yudkowsky & Nate Soares”?
Plus someone with a 17-character name like “Eliezer Yudkowsky” simply can’t have such a large author font. You’re spending three lines of text on the author names!
Plus I would understand making the author names so large if you had a humungous pre-existing readership (when you’re Stephen King or J. K. Rowling, the title of your book is irrelevant). But even Yudkowsky doesn’t have that, and Nate certainly doesn’t. So why not make the author names smaller, and let the title speak for itself?
I understand the artistic desire to have the irrecoverable red event horizon of superintelligence to underscore the “would kill us all” part, but since it makes the words “kill us all” harder to read, I’m not sure if this current design underscores or obscures that.
Surely it would’ve been possible to somehow make title and subtitle less than four lines of text each?
And overall, the cover just looks very cheap and low-effort.
EDIT: More fundamentally: in all media, title & cover art are almost as important as content, because you can’t get people to click on your video, or to pick up your book in a bookstore, if title & cover aren’t good. Then it doesn’t matter how great the content is, if nobody ever sees it.
Anyway, the title is good, the cover is bad, and I can’t assess the content yet. You say this book has been in the works for over a year, and that you spent lots of effort on polishing its content. Then if it’s a good fraction of MIRI’s output for that time, and e.g. a cover is responsible for (say) 20% of a book’s impact, then wouldn’t that justify spending >>>$100k on the cover design? This one looks more like a cover purchased on Fiverr.
Also see this discussion on the need to spend a significant fraction of one’s effort on a piece of content, on stuff like title and cover and thumbnail and book blurb.
The “lightcone-eating” effect on the website is quite cool. The immediate obvious idea is to have that as a background and write the title inside the black area.
If one wanted to be cute you could even make the expansion vaguely skull-shaped; perhaps like so?
Most of those are fiction or biographies/memoirs (which often have a picture of the subject/author on the cover), which seem to have a different cover style than other books. Skimming through some lists of NYT bestsellers, some books with the most comparable “Really Big Thing!” topics are “Fascism: A Warning” (Madeleine Albright, cover has large red-on-black lettering, no imagery), “How to Avoid a Climate Disaster” (Bill Gates, cover has large gradiented blue-to-red text on white background, author above, subtitle below, no imagery), “Germs” (title in centered large black lettering, subtitle “Biological Weapons and America’s Secret War” in smaller text above, authors beneath; background is a white surface with a diagonally-oriented glass slide on it), and “A Warning—Anonymous” (plain black text on white background, subtitle “A Senior Trump Administration Official” in small red lettering below, no imagery). Neither cover version of IABIED looks that different from that pattern, I think.
Given that the book is being published by a major publisher, it can safely be assumed that the cover design was made by a professional cover designer, who knew what they were doing.
Contrary to what you wrote, the title has a bigger font size than both the subtitle and the authors’ names (this is true of both the American and UK covers; I am primarily talking about the American cover, which I presume is the one you are referencing). Even if the author names were the same size as the title, it is immediately obvious which one is the title, and which one isn’t. Putting the subtitle in a dark grey, which is much closer to the background color (black) than the color of the title (white) is, also does a lot to move emphasis towards the title of the book (away from the subtitle)
Most importantly, the title is plenty big. If it was small, then I would feel there is something to what you are saying; but the title is quite large and readable from a distance, and clearly delineated from the rest of the text on the cover.
In this case, part of the point of publishing a book (including writing it in the first place), is presumably to promote the identity of the authors, to make them a known name / schelling point for discussion about AI safety. That would indicate making the names quite prominent on the cover.
I see that the numbers indicate people disagree with this post. Since there are several clauses, it’s hard to know which specifically (or all of them) are being disagreed with.
The second paragraph (beginning “Contrary to what you wrote...”) is a list of factual statements, which as far as I can tell are all correct.
The third paragraph (“Most importantly, the title is plenty big...”) is more subjective, but I’m currently not imagining that anyone is disagreeing with that paragraph (that is, that anyone thinks “actually, the title is too small”).
The fourth paragraph (“In this case, part of the point...”) is more speculative, and I could easily imagine someone reading it and thinking “that’s not the point of publishing / writing a book”. There’s certainly a reason I put a “presumably” in there. I do still feel that there’s something to what I’m saying in that paragraph. My surprise would be of a limited extent if Soares and Yudkowsky said “that was not a consideration in our decision to do this”—but I would be somewhat surprised.
I can see someone disagreeing with the first paragraph (“Given that the book...”), but my current state of mind is that such people would be simply wrong. The book is not being self-published, but is being published by Little, Brown and Company. Some excerpts from Wikipedia’s article on Little, Brown and Company:
and
The point being, the company that is publishing Soares and Yudkowsky’s book, is an established company that has sold important and/or bestselling works for two centuries. The people there know what they are doing, and that includes the people who design covers, as well as the bosses of the people who design the covers.
I imagine most disagreement comes from the first paragraph.
The problem with assuming that since the publisher is famous their design is necessarily good is that even huge companies make much worse baffling design decisions all the time, and in this case one can directly see the design and know that it’s not great – the weak outside-view evidence that prestigious companies usually do good work doesn’t move this very much.
Yes, my disagreement was mostly with the first paragraph, which read to me like “who are you going to believe, the expert or your own lying eyes”. I’m not an expert, but I do have a sense of aesthetics, that sense of aesthetics says the cover looks bad, and many others agree. I don’t care if the cover was designed by a professional; to shift my opinion as a layperson, I would need evidence that the cover is well-received by many more people than dislike it, plus A/B tests of alternative covers that show it can’t be easily improved upon.
That said, I also disagreed somewhat with the fourth paragraph, because when it comes to AI Safety, MIRI really needs no introduction or promotion of their authors. They’re well-known, the labs just ignore their claim that “if anyone builds it, everyone dies”.
I used to do graphic design professionally, and I definitely agree the cover needs some work.
I put together a few quick concepts, just to explore some possible alternate directions they could take it:
https://i.imgur.com/zhnVELh.png
https://i.imgur.com/OqouN9V.png
https://i.imgur.com/Shyezh1.png
These aren’t really finished quality either, but the authors should feel free to borrow and expand on any ideas they like if they decide to do a redesign.
It’s important that the cover not make the book look like fiction, which I think these do. The difference in style is good to keep in mind.
Those are definitely all improvements on the current cover!
I only like the first one more than the current cover, and I think then not by all that much. I do think this is the sort of thing that’s relatively easy to focus group / get data on, and the right strategy is probably something that appeals to airport book buyers instead of LessWrongers.
Finally created a LW account (after years of lurking) to upvote and agree on the cover design issue.
This is a topic I read whatever I can get my hands on, and if I saw this book in a store (and did not know EY or Nate), even I would be a bit put off to give it a read
Given the stated goal of trying to make this a bestseller, I feel like the cover is a pretty big impediment