IMX, the toughest to find errors for an author or the proofreaders are typos of the kind "is / in", "or / of / on", "is is", "the / he", missing plural-s etc. As a rule, these typos are noticed while the book is already being printed;) Every book has a few such typos. If it has many, then a spellchecker was not used and proofreading was maybe absent. Of course, what matters is not typos but is contents.

Some books have awful errors, such as wrong diagrams, and everybody would notice this. Of the remaining errors of the contents, a part is easy to find by every reader, because the context is easy.

Other errors are hard to find if the contents is advanced and detecting the errors requires a pretty advanced understanding. Although stronger players can detect such errors more easily on average, there can be errors that remain hidden to the reader or even to everybody for many years. Possibly even to the author. From my own experience as a reader, Counting Liberties and Whole Board Thinking are examples of books, for which I started to discover mistake(s) in the contents only several years after reading. I could discover errors only after I studied related topics for myself carefully and systematically. So, to find (preferably all) important errors in a book, the reader's strength or experience as a player alone might not be enough.

As an author, I proceed pragmatically: the first proofreading reveals many errors, the second one or two dozens, the third a handful to a dozen and the fourth about 2 or 3. During the process, too difficult contents is dropped entirely, so that I am not confronted with contents above my own understanding within reasonable time. Excluding even the last error is necessary only for books with declared 100% correct contents. Otherwise, it can be tolerated that a book has 0 or 1 remaining errors in the contents. Now, which errors should the readers detect at all? Those that do not exist? Whole Board Thinking is such a book by another author, which might have exactly 1 error of the contents.

As a reader, I find many errors quickly, if a book has many errors, and I am already very familiar with the topics, such as The Theory and Practice of Analysis. As a 10 kyu, I would have overlooked (almost) all of the errors.

So my preliminary conclusion is: finding errors is the easier the more errors a book contains, the more prior understanding the reader has already had about the topics, the stronger the reader is and the less advanced the theory or concept of the book is. Are there further important factors?

I guess a native speaker would correct every sentence in the above paragraph. Not for typos or errors but for language. To me it shows that you ( and I if time comes ) need to hire a native speaker to proofread before publishing in a foreign language. Second advise is to proofread from paper not from screen. Third advise is not to proofread your own texts. Final advise is not to proofread more then say 10 pages a day slowly reading every sentence aloud. Even then 1% of the errors will be missed.

_________________
I think I am so I think I am.

The third one was ok.

_________________
Help make L19 more organized. Make an index: https://lifein19x19.com/viewtopic.php?f=14&t=5207

Native speakers overlook such, too.



Basic.



It helps a lot if both the author and others proofread; more proofreaders of the same text find more.



Good in theory, but unrealistic.

The ultimate tip I got for proofreading was reading the text backward, word by word. Meaning you don't primarily read sentences through experience/recognition because reading in this order only yields gibberish. Takes ages though and mainly helps to spot spelling mistakes.

_________________
My "guide" to become stronger in Go