Yes, don’t judge a book by its cover - but lots of people do anyway.
If the AI produced something indistinguishable from a good stock photo, then no, I would probably not have noticed. But it didn’t, and they still put it on top of their hard work. If the AI image sucks, don’t use it!
a tasteful public domain image of a crab or something like hundreds of years old but then detailed enough kind of like one of those O'Reilly's books and then make the whole cover white with some text that stands out... It has all been done before and I anal but I doubt O'Reilly's can copyright or trademark that design feeling.
Edit: unless the authors are trying to crowd source the proof reading and peer review by releasing an early access copy? This cover draws attention from the average reader who is drawn to comment about the book because there is something to criticize? Maybe this is some kind of advanced 3D chess?
Choosing to make your book "fit" is often a good call. You're actually going to attract people who know roughly what they want. My mother reads a lot of trash romance, it's not very important who wrote it, but a handsome guy on the cover signals that this is roughly the right book, the attire and surroundings are a vague gesture at the sub-genre, guy in a white coat? Medical. Business suit? Maybe 1980s.
There's no rule against refusing to obey convention, "Twenty Jazz Funk Greats" is, in fact, not Jazz-funk, and indeed didn't have twenty tracks. But nobody working on that album thought "This is a commercial route to success". It's art. Really difficult to listen to art. If you are not making art but have some other purpose you probably do want to signal your intent to people merely browsing.
For example, here are some images from Library of Congress with crabs in them that I think could make for interesting book covers:
- https://www.loc.gov/pictures/resource/fsa.8a23704/ "Residents of Raceland, Louisiana, eating crabs at crab boil" (1938 Sept.)
- https://www.loc.gov/pictures/resource/jpd.00748/ "Benkeigani to tsubaki" (between 1825 and 1830)
- https://www.loc.gov/pictures/resource/ppmsca.69838/ "Frozen crabs at a restaurant" (January 1954)
- https://www.loc.gov/pictures/resource/highsm.14204/ "The Barking Crab restaurant on the Fan Pier across the Northern Bridge from downtown Boston, Massachusetts" (between 1980 and 2006)
- https://www.loc.gov/pictures/resource/ds.00242/ "Cancer" (latin name for crab, you'll probably want to edit the image to avoid any unfortunate connotations) (1825)
But before picking any of these or other images from the Library of Congress, check the "about this item" link and see in particular the "Rights Advisory", including any links to further details about rights and usage in that field, and keep also in mind things like what is pointed out in one of those links; "Privacy and publicity rights may also apply". IANAL, TINLA.
Especially with an ebook, if the content is good, I frankly don’t even care if it even has a cover at all.
That there's a fixed budget of time/money - exactly X hours/dollars will be allocated to the book, one way or another.
That the improvement to the quality of the content as time/money is invested is something approaching linear - that if exactly X hours/dollars are going to be allocated to the book, the content will be appreciably better if, say, 0.9X is allocated to the content vs, say, 0.8X.
That time and money is fungible: that all time/money that could be invested in the content could be invested in superficial things, and the other way around.
As independent developer.
Yes, time is fixed. There is only 24 hours in a day.
For example, if you were to ask a human artist to draw this image with two crabs, and this artist didn't know how many legs crabs have, what would they do? I think they'd probably google how many legs crabs have, and how they look in general, and at the very least draw two crabs with the same amount of legs.
But these generative models don't know how many legs crabs have, and they are not clever enough to ask, or even count it seems. So you end up with results like these. The main crab has 8 legs (6 + 2 claws), while the little one seemingly has 10 (or at least it has 5 on one side), and it's also missing an eye for some reason.
A human artist can also make mistakes, for sure. But i don't think these kind of mistakes would be expected from an artist.
And i agree with the other commenters that said this reflects poorly on the author of the book. Didn't they care to check the cover for obvious mistakes? Attention to detail is a very important trait on a technical writer.
This isn't an accident. Publishers put a lot of effort into ensuring covers send accurate signals about the contents of the book. A book with a garbage cover didn't come through the modern publishing process, and it didn't get made by someone who cares enough about their book to put actual effort into the cover. The percentage of books that will be worthwhile with a bad cover is vanishingly small. To a first approximation, it's 0%.
It's very reasonable to judge a book by its cover these days. It's an intentional signal; covers aren't chosen at random. If you intend your book to be taken seriously, you need to keep up with modern standards.
In sociology most of the results that actually replicate are the ones that are true but people wish they weren't.
Even worse, it makes me question the content itself. If they used Stable Diffusion for the cover image, who's to say they didn't also use an LLM to generate some of the Rust examples in the book itself?
Note: I'm not suggesting that the inherent usage of an LLM to generate Rust code is necessarily bad. What I am suggesting is that they might approach the review of said generated code with the same laissez-faire attitude as they did with the cover image.
It seem to also be missing left legs, as if it was damaged.
The book seems focused on errors and error handling, maybe a 'bug' on the cover was on purpose.
Not so sure this is just AI artifact.
a huge amount of the comments on the book were critical just because of the cover.