This post is an attempt to reach out to publishers. So that they hear and take their translations more seriously.
Throughout my career as a developer, I have bought many different books. Books from various publishers, both small and large. Primarily from large publishing houses that have the resources to invest in translating technical literature. These were diverse books: we have all gone or are going through the path of self-discovery. And what connected all these books was one thing: they were translated in a way that made them unreadable. Over time, of course, you get used to the translation of terms (mentally translating them into those that are used in everyday life) and the broken writing style, which makes it clear that this text has been taken from English. However, one never gets accustomed to the price that publishers ask for popular editions.
Publishers are invited to comment.
Let's try to understand, what is a book? Let's take a 600-page book that is an average offering in the IT publishing market. The cost of printing one copy, based on the prices of Chekhov Publishing House, used by large publishers, is 175 rubles. And printing, for example, 2,000 copies costs 350,000 rubles. Furthermore, if we take a popular book, its price will be around 1,500 rubles. That is, the publishing house will receive (1,500 - 175) * 2,000 - 13% = 2,305,000 rubles.
But the publishing house has many expenses. Below are my pathetic attempts to calculate, but in the comments, the Publisher Peter came and explained it in more detail. I'm copying from the comment + link to the comment:
My pitiful attempts
- to pay for the warehouse;
- for transportation from the printing house to the warehouse;
- distributor services (as far as I know, about 150 rubles per book… but that’s a fantasy)
- translator and editor services;
- a small percentage — salaries of the entire publishing team (there are many books, hence the percentage is small);
Answer . There’s also , I recommend reading
- You forgot about the payment to the rights holder/author (advance + royalties).
- You incorrectly calculated the taxes (underestimating them). There is VAT, and there are actual taxes.
- You have not considered the "turnover rate," which dictates the requirements for margins. As you have noted, a book is not published in a month. The print run does not sell in a month either. And the costs from the very start are quite substantial (advance + administration, which precedes the search, acceptance for publication, and obtaining rights). These expenses accompany the book until the last copy is sold. If the edition does not generate more income than alternative investment methods, what is the purpose of the publisher's existence?
- If you have a team, then you have an office (or offices) where they work on certain computers, etc. Maintaining them costs money.
- The assumption that employee salaries are a small percentage is only relevant if there are indeed many books. But if there are many, they inevitably receive less attention (which you do not like). And if there are few books in progress, the percentage of these expenses cannot be small. In general, this expense category dynamically occupies as much as readers are willing to pay additionally for it.
- Commercial risk. Not all books sell within the planned timeframe, which means that at best not all books are profitable. Moreover, not all books sell out, in principle. Naturally, all these risks are calculated and compensated by increasing the prices of all upcoming books. Thus, popular books cover the losses of unsuccessful ones.
- The most flawed point in your calculation is the distributor's commission. It is not a fixed 150 rubles. It is not fixed at all. The publishing house ships books in bulk. Retail chains place them on shelves at any price they deem justified. Your calculation indicates an increase in the publisher's price of about 10%. This is very far from the truth (the difference is substantial; the increase from the publisher's price can reach 60%, which the wholesaler keeps for themselves).
Therefore, there will be some output, but it won't be magical. For instance, accounts will ultimately receive just over 500,000 rubles from 2,000 copies. From the perspective of large businesses, this amount is not exactly serious. Hence, publishers begin to cut costs. For instance, in the list above, I did not mention proofreading by technology specialists related to the books written. Why? Because publishers have a model where 'technical specialists proofread the book for free, correct it, and in return, get their name printed in small text somewhere that nobody reads.' For some, it's a sense of self-importance; for others, it's a way to reduce expenses. It sounds great, if not for one 'but.'
Publishers do not need our edits.
Not everyone knows, but I have a , which I write from time to time. It is hosted on GitHub and distributed under a free license. I approached two publications with this work (I won't mention names, but their books are on your shelves). At first, I tried to reach out in the early days when it was about 30% done. After a lengthy correspondence (about 80 emails), we debated:
- I wanted my cover, which I ordered from a designer at Lebedev Studio. They said no;
- they wanted me to remove all copies of the book from GitHub. This is impossible, so I argued that it was impossible;
- I wanted to retain the right to publish an English version separately. They imposed a ban, justifying it by saying that if an English-language publisher approached them, they did not want to give up the opportunity to profit from it. But no one has ever approached them yet..
I demanded to amend the contract, but they arranged it so that externally it appeared as though I could publish in English separately — at another publisher. But in reality — no. That ended the conversation.
I approached another publication. They requested to review the text, which I sent. They laid out the terms:
- Publishing would cost me from 200,000 rubles.
- Starting from 500 copies.
- Low-density paper (similar to newsprint, where the letters bleed through);
- In sales — 45% for me, 55% for them.
At the same time, the work was checked by their translator. So, what does this mean?
The publishing house does not have programmers. Instead, there are people who specialize in technical translation. The management doesn't have programmers. What does this indicate? It shows that the management doesn't understand the content of the text. Essentially, they are only concerned about sales. There is a person on staff who translates technical literature. Surely, he has become quite experienced in this? This means they trust him, considering him an expert in the field. This person receives a book from some author and compares it with his own experience. Since he receives books in a continuous stream, plus some are in process, he won’t deeply engage with the text. What has been written to me:
Quote:» This is not at all a destructor, as may initially be assumed due to the similarity of finalizer declarations in C# and destructors in C++. A finalizer, unlike a destructor, will be called guaranteed, whereas a destructor may not be called at all.
Translator: The statement "a destructor in C++ may not be called" is complete nonsense (not to mention the awkward use of the verb form here).
The discussion of exceptions in the second part is more interesting, but hardly original — all of this is surely present in Richter's book 'CLR via C#'. The promised multithreading is well covered in the translated <by the publisher> book on the subject.
The author's handling of terminology also does not contribute to the book's credibility.
But here is another example: literally on one page, three translations of one term (stack unwinding) are used: unwinding, unwinding, and unraveling. How is that to be assessed?
In general, for publication in book form, the material needs either to be rewritten by you or carefully edited.
I do not claim to have good style or to be free from grammar and spelling errors. But… does the translator analyze the technology description errors? Moreover, confidently proposing everything be rewritten without considering that he might not know something. The response was the following:
if the memory is not freed from the object, the destructor will not be called, as there will be a memory leak.
Exceptions are superficially described everywhere, unlike in my book.
The author's handling of terminology also does not contribute to the book's credibility.
This is the terminology of programmers. Is your expert a developer on the .NET platform?
But here is another example: literally on one page, three translations of one term (stack unwinding) are used: unwinding, unwinding, and unraveling. How is that to be assessed?
All three words are actively used.
At the same time, I tried my hand at editing translations from English to Russian. The text is a typical mess. Both in style and in the translation of terms. In other words, it’s written in Russian, but not in a Russian way. It sounds English. Familiar? I roll up my sleeves and start editing. Sometimes, whole paragraphs. The response was roughly: why are you doing this? We know better how it should be done. Our translator is very good, and after him, there’s no need to look at the style and translation. Only some terms, code listings. No need to waste time on translation.
How it should be
For English translations, I use . He and his team have a completely different approach. So I have something to compare. Initially, he and a second translator bombarded me with questions. About inheritance, virtual method tables, about GC. They asked so many questions that I’m sure they both would pass an interview for a .NET programmer. Over time, the number of questions decreased significantly. And at this point, there are almost none. Why? Because they arrived at the correct terminology. Recently, he sent me this:
To say I was surprised would be an understatement. So it turns out that translations can be good? 🙂 But with one condition: when the editing from the programmer goes parallel with the translation, rather than at the very end, when the publishing house regrets the time spent.
The editor and the reviewing programmer must work in parallel with the translation
Conclusions for myself
Publishing houses don’t need high-quality translations into Russian. It’s too expensive for them. While the programmer proofreads, while the full editing is done, while it is agreed upon with the publishing house (disputes over every paragraph), a lot of time passes. It could even take a year. During that time, the technology may become outdated and unnecessary. And the book should be shelved right now while the topic is hot.
On the other hand, the internet is full of articles. Free articles. And the publishing house is losing customers. Especially with a poor translation. But, esteemed publishing houses, why do we buy books?
Personally, I prefer books because the author thinks more globally compared to the author of an article. That is, I get a deeper and more thoughtful description of the technology. I find it easier to read a book than on a reader or screen. There's no screen light, and I can flip through the pages. I get tired of screens and crave something tactile. A book.
So, dear publishers. The mammoths of the printing industry. There is a standard for translation among translators. If the initial language is translated by a native speaker, then the editing is always done by a native speaker of the target language. This doesn’t seem strange to you, does it? It’s logical and you consider it normal. In the case of IT books, the native speakers are programmers. We must be listened to. So that we can read your books later, and you can earn income in the era of blogs and free information.
Only registered users can participate in the survey. , please.
Technical book translation:
I still take translations
I haven't read translated books for a year now
I haven't read translated books for two years now
I haven't read translated books for four years now
I haven't read translated books for over five years
175 users voted. 46 users abstained.
About editing
We need to listen to programmer-editors, trusting them. Checking, but trusting.
Translators do well, but programmers are not writers, and it’s better not to listen to them.
Your option (in the comments)
133 users voted. 52 users abstained.
Source: habr.com
