Bombora. Bloody and merciless translation of books
You buy a paper book, inhale the aroma of its pages, start reading this miracle of human thought and the printing press, when suddenly… #%^#@&*#^~$%? Exhaling a little, you are trying to understand what, in fact, for the nonsense you just uploaded to your brain? In this article I will share my experience gained from trying to read a translated book from Bombora publishing house. I just warn you right away: from the examples of “translation” shown, hysterical fits can happen.
What’s happened
In the summer of this (2023) year, I found myself in a bookstore where, as many of you understand, it is extremely difficult to resist and not buy yourself some useful book. I am a programmer, so an unknown force led me to the counter with computer literature.
At that time, there was only one book related to development in my home library – “CLR via C#. Programming on the Microsoft .NET Framework 4.5 in C#by Jeffrey Richter. And she urgently needed a friend to stay in the closet drawer!
Having carefully examined the rack, I opted for the book “Algorithms for beginners. Theory and practice for the developer“. I have not been a beginner for a long time, but I wanted to have a small reference book on the most common algorithms at hand, refresh my knowledge, and, in the end, find something new for myself (and it is there). The copy was bought for 1400 rubles, which is generally a normal price for a technical book in a bookstore, but, as you will see below, it is slightly overpriced. Approximately 1400 times.
My heart was beating wildly, my legs gave way, the anticipation of an exciting reading was driving me crazy. And then the day and hour arrived. Having reached almost the end of the preface, I stumbled over the phrase:
Chapter 14 describes classification of algorithms: the algorithm learns to group data, …
“Describes the classification of algorithms: an algorithm learns to group…” A? Maybe I’m already tired, just starting to read, but there is clearly some kind of inconsistency. After reading the sentence again with the mode of maximum thoughtfulness turned on, the feeling of absurdity did not go away. Perhaps you mean the classification algorithm? There was an idea to look into the original text in English:
Chapter 14 covers a classification algorithm: the algorithm learns how to classify data, …
And it’s true. That is, I should have seen in the book in front of me this phrase: “Chapter 14 describes the classification algorithm“.
Okay, almost every book has typos and a little bit of nonsense. But page 28 sowed doubts about the adequacy of the upcoming chapters:
Take a look at Figure 1.2, which shows two functions: f1(n) = 20n + 1000 and f2(n) = n2. For small values of n, this is f1(n), which takes the largest values, but the situation changes very rapidly, after which n2 grows much faster.
On the graph for f2(n) you will of course see a parabola. But even without a schedule, each of you has already noticed nonsense, and even repeated twice – n2. Strange things are happening with indexes in this kingdom. The original is expectedly clear in front of its readers:
Take a look at figure 1.2, where we plot two functions, f1(n) = 20n + 1000 and f2(n) = n2. For small values of n it is f1(n) that takes the biggest values, but the situation changes drastically quite early, after which n2 grows much faster.
And immediately, let’s look at the next paragraph in the original and Bombora’s translation. Original:
The big-Oh notation allows us to simplify functions. If we have a function like f(n) = 3n3 + 5n2 + 2n + 1000, then we have simply O(f(n)) = n3.
Translation:
Big O notation allows us to simplify functions. If we have a function f(n) = 3n3 + 5n2 + 2n + 1000, then we simplify it to O(f(n)) = n2.
Well, the index finally went up. But one does not need to be an expert in estimating the complexity of algorithms to understand that a large O for a polynomial function of the third degree will be a cubic function, and not a quadratic one. The funny thing is that the next short sentence is: “Why?“. Eh, I wish I knew…
In fact, the state of light fucking overtook me earlier, from the very first chapter. My emotions are better conveyed review to the book on Ozon:
A disgusting, shockingly thoughtless translation. It feels like it hasn’t been read at all. In the very first chapter, the first paragraph: “The difference in the value of shares on a particular day is the number of days following each other, from the one we have chosen and in the opposite direction …”. What? What did I just read??? The difference in the value of shares is the number of days??? Look at the original: “The span of the stock’s price on a given day is the number of consecutive days..” . And the author of the translation drags this price difference throughout the chapter, due to which common sense is completely lost. That is, even the title of the chapter is translated incorrectly.
Unfortunately, I didn’t read any reviews before buying and didn’t even know about the existence of the book, as well as the Bombora publishing house. By the way, how often do you see a 3.6 rating for products? There are other comments there, read if you’re interested. For a longer list of in-book absurdities, see later in the article in the section That’s what everyone’s come for.
The rules of good manners tell us to first write about mistakes and shortcomings personally. I respect these rules, and therefore on August 19 I sent a letter to the publishing house about the cruel treatment of the text of a foreign author. Email bombora@eksmo.ru taken from the official site, there is even a postscript “Found a mistake on the website or in the book? Doubt and do not know where to write? Write to us in any unclear situation.”. Here I also wrote. As of August 27, there is no response. In this case, I think it is right to voice the problem to a wider audience.
Bombora and their translations
As I said, I was not aware of the publishing house. However, starting to study information about him, curious points came to light. Firstly, this is a division of the very famous Eksmo. And secondly, in the category of non-fiction, this is one of the leading publishing houses. Therefore, the office is serious, from which you expect an appropriate approach when working with books.
In 2018, the DTF website published a devastating article about how the guys from Bombora did not spare the work of Jason Schreier and released Blood, Sweat and Pixels. The material became resonant so much that the publishing house released new version translation. And in December 2020, it was held event “How to use bad translation and hatred of the fan community to your advantage by building a new and well-coordinated workat”. I don’t know how the construction was carried out, but my experience indicates the lack of work on the bugs.
But even earlier, in 2019, the publishing house of Bombora already talked about the difficulties of translation. Quote from the material at the link provided:
To reduce such situations to a minimum, the BOMBORA publishing house launched the “We Translate Books, Not Arrows” program, within the framework of which a double check of the translation is carried out, …
Just memorize the phrase about double checking the translation and keep it in mind as you read this section. That’s what everyone’s come forit will be fun.
Do not think that I am writing this article with bad intentions and slander the publisher. Yes, everyone makes mistakes, the ability to admit them and become better deserves great respect. But it is one thing to say that the lessons have been learned, and another thing to translate like Bombora. I will give some examples of people’s reactions to their books.
Bombora are those who make a bad translation? Or is it others?
Bombora is a black mark. They have beautiful covers and interesting books in the original. But the translation is machine-made, without any minimal editing. And they also have a feature – to republish books, but with a “new translation” die. Like they hint that they know what the hell they released in the 1st edition. But they sold anyway.
I bought a book about Miyazaki – I have never seen such a terrible translation. There is simply no editing. Even minimal. Literally every sentence is clumsy.
For the first time in my life, I asked what kind of publisher it was. Bombora. I will never buy anything from them.
Hate Clock: ‘Bombora’ criticized for translating Friends book
The birth of machines. Unknown history of cybernetics | Reid Thomas
The translation leaves much to be desired. Without knowledge of English and assumptions about what was written in the original, the meaning of some sentences may be lost.
The evolution of the mind | Kurzweil Ray
I gave it to a good friend to read. He said that the translation is bad and there are many typos. How could they entrust the translation of such an important book to someone I don’t understand? Bombora (Eksmo) seems to be a respected publishing house, I had their encyclopedias as a child, and therefore it’s a shame that this is so.
And this, of course, is an incomplete list.
IN interview In 2019, the editor-in-chief of the publishing house Ramil Faskhutdinov announced a plan to earn 3.5 billion rubles for that year. Probably, the revenue covers the costs in the form of negative reviews. Although, it seems to me that by doing well and ridding yourself of the wrath of the masses, you can earn even more without losing reputation points.
Here is a quote from another interview:
One bad translation gets more attention than a thousand good ones. The problem of bad translations is not in the realm of bad translators in applied books. The specialist text itself can be dry. In English, the perception of the text is slightly different. Someone translates literally, someone – literary changing. The editor should check. I can’t know if the editor checked well or not.
That is, the author of the original is to blame, the editor cannot do a-ta-ta in the publishing house, we are not to blame, of course. I agree that when something works well, no one notices it, and when it doesn’t work, people start to resent with a pitchfork in their hands. But above you can see that there is a problem, it is unprofessional to look for excuses. The first time was an accident, the second time it was a coincidence, the third time it was a pattern.
That’s what everyone’s come for
Before I show examples of translation, the problems of which, of course, are “is in applied books not in the realm of bad translators”, it is worth mentioning the editors of the book “Algorithms for beginners. Theory and practice for the developer”:
Editorial Director: E. Kapiev.
Managing editor: E. Istomina.
Junior editor: E. Minina.
Three editors and a myriad of nonsense in the text. Yes, it happens.
General
No one in the professional environment uses the terms “input/output data“. For everyone it’sinput/output data“. All the algorithms in the book suffered.
Page 53
(b) Directed cyclic graph without loops
Image with this caption:
In original:
(b) A directed cyclic graph with loops.
Page 56
In undirected graphs, the set E is the set consisting of two sets of elements {x, y} for each edge…
You don’t even need to open the original. Should be “set consisting of two-element sets (sets, each of which has 2 elements)”, we are talking about the list of graph edges:
In undirected graphs, the set E is a set consisting of two element sets {x,y} for each edge …
And this nonsense drags on further down the text.
Page 57
Each element of the array represents one of the vertices and starts a tupleconsisting of vertices, …
Runs a tuple, you say? Not otherwise than in space:
Each element of the array stands for one vertex and is the start of a list that contains the vertices
Should be something like “Each array element represents one vertex and is a pointer to a list containing the vertices…“.
Page 70
Initial a recursive function is a factorial, …
Initial? And what is the second one? And the third? Let’s look at the original:
The archetypical recursive function is the factorial …
The correct translation is something like this:The classic example of a recursive function is the factorial“.
Page 118
We are dealing with an example stalemate.
Original “corner case” This “boundary/edge case“. There is nothing dead-end about him, he is just special for one reason or another.
Page 212
The original has this example:
Bombora thought it was a good idea to transfer him. And they identified an important requirement: do not bring the text of the chapter itself into line with the translated lines! What’s in the original
The method broke lines when their length approached thirty characters (including spaces and punctuation).
what’s in translation
This technique breaks the lines as soon as their volume reaches thirty characters (including spaces and punctuation).
After all, the lengths of words and sentences in Russian and English do not differ, right? How about Figure 7.3:
That is, examples and figures are simply not connected with the main text. Why do you need to translate examples at all, I sincerely do not understand.
Page 218
A tree whose nodes are the nodes of the graph and whose edges are a subset of the graph’s edges is called spanning tree.
original
spanning tree
in Russian-language graph theory it is “spanning tree“.
Page 219
The algorithm uses minimum priority queue…
Original:
The algorithm uses a minimum priority queue …
Correct translation: “The algorithm uses the minimum priority queue…“.
Page 220
The loop in lines 11–17 is repeated until the priority queue is empty (line 11), it takes the smallest element out of it (line 12), and relaxes estimated path data…
How do you relax? Original:
It extracts the minimum element from it (line 12), and relaxes the path estimate …
About the termrelaxation” not only in graph algorithms, but also in other technical sciences, the entire team of editors, together with the translator, do not know.
Page 302
A key may contain more than one attribute. In this case, we are talking about linkeror a composite key.
Original:
A key may be made by more than one attribute. When this happens, we call it a composite or a compound key.
The composite key has become a linker, so-and-so.
Page 310
You can see the result in figure 11.9. Almost a straight line consideringthat Benford’s law is related to power laws.
Everything seems to be fine, however, the cause-and-effect relationships are turned inside out. Original source:
The result is in figure 11.9. That is almost a straight line suggesting that Benford law is related to power laws.
Correct translation: “An almost straight line, indicating that Benford’s law is related to power laws“. That is, instead of being surprised by the unexpectedly straight line of Bombora in the original, this line is evidence of a certain distribution of data. Even Yandex translator gives adequate text.
Page 316
Here the author of the translation decided to make a complete adaptation for the Russian-speaking audience. The original is about Secretary problemwhich we know as The picky bride problem, and in this form the formulation of the problem went into Bombora’s book. Okay, so be it. But bad luck, the name of the function in the algorithm is this:
Algorithm 11.5. bride search
SecretarySearch
Perhaps the bride works as a secretary. Then everything is fine.
Page 439
To resolve this nuisance, we need to abandon the use of a bitmap as a hash table, and look towards a countable array, such as an array of zero-based integers.
This is the height of insanity. Original:
The solution is not to use an array of bits as the hash table, but an array of counters, for example, integers, initialized to zero.
The original thought was:The solution would be to use as a hash table not a bit array, but an array of counters, for example integers, initially set to 0“.
Page 449
The I Ching (易經), or the Book of Changes, is an ancient sacred Chinese text that appeared between the 10th and 4th centuries BC. Process divination The I Ching includes…
Why did you suddenly talk about some kind of fortune-telling process? No one told us that this text had anything to do with him. Everything becomes clear if you look at the source:
The I Ching (易經), or the Classic of Changes, also known as the Book of Changes, is an ancient Chinese divination text. It was probably assembled some time between the 10th and the 4th centuries BCE. The process of divination in I Ching involves …
Oh yes, in the original, in the very first sentence, we were told that the text was for predictions (and not sacred at all). A little further on, I finished reading, for there was no longer any desire to torture my eyes and brain.
aftertaste
The list above is far from complete, I do not have enough internal strength to collect all the diamonds of a poor-quality translation. Yes, and it smacks of masochism. My book is full of notes and corrections. I decided that since I bought the book, even with barbaric proofreading with an ordinary ballpoint pen, it would contain meaningful information.
Of course, the text is full of spelling errors. I don’t pay much attention to them, they don’t change the meaning. But semantic errors are a terrible thing. The book is intended specifically for beginners, it does not even have a section on dynamic programming. What will be the reaction of a person who, unlike me, is really taking his first steps in the topic? Perhaps he will understand that the matter is in translation. And perhaps he will consider the author of the original ignorant.
Non-fiction and even more so technical literature has an educational function. The book of this segment should be convenient and pleasant to read, it should provide correct and meaningful information. “Algorithms for beginners. Theory and practice for the developer” not only does not cope with its task, but also encourages two steps:
Read technical texts only in the original.
Never mess with Bombora products.
I believe that the book fulfilled its purpose from the point of view of the management of the publishing house – it brought in revenue. Is it worth the loss of respect? I think Bombora has given itself the answer a long time ago. In confirmation of this, silence on my letter to them. There was not even an on-duty response like “We are sorry that this happened.“. I’m not even talking about republishing a book with a normal translation and providing it for free to the owners of the current one.
And I don’t even need that. Just start taking a more responsible approach to adapting foreign texts, they carry knowledge and useful information that you simply kill with your translations. Start loving your readers and readers will love you.
From section About publishing house:
Bombora is the dream wave. For Australian surfers, it is akin to a mystical sign: if you catch a bombora, then your life will never be the same.
The book market is an ocean. And every reader dreams of catching a wave that will turn his life upside down. Your BOMBOR.
That’s for sure, life will not be the same, I will choose translated books more carefully. Fortunately, there are publishing houses that care about their image.
