Posts Tagged ‘translation’

From the Norwegian Association of Literary Translators:

Try to ask this question to several stakeholders and you would be surprised about how many different answers you will get.

Some may say that quality is measured by the user experience and that a quality localized product is one that functions the same way as the English version.

Others may say that a quality translation is one that maintains brand consistency.

Or that a quality translation is one that is factually accurate, readable and (hear hear) not localized (preserves the source culture nuances).

I find all the above answers valid.

However, in June 2006 a new translation quality assurance standard was published by ASTM International and unfortunately it’s still relatively unknown: ASTM F2575.

The ASTM translation standard (F2575) defines translation quality as:

The degree to which the characteristics of a translation fulfill the requirements of the agreed upon specifications

This definition implies two stages: Agreeing upon project-specific specifications and applying those specifications. Sounds too easy, doesn’t it? But it actually works! This approach can be applied to every translation project.

How? Well, translation projects usually consist of three phases:

Pre-Production, Production and Post-Production (aka Post Mortem).

It’s in the Pre-production phase when you should discuss and agree upon the specifications. In the Production phase, these specification should be applied. Finally, in the Post-Production phase, you should carry out the project analysis to verify the fulfillments of the agreed upon specifications.

If all translation projects followed this simple approach, all the different stakeholders would be much happier at the end of the project!

According to an article published in Bloomberg’s Businessweek on 10/15/2012, based on calculations of the Bureau of Labor Statistics data, translation and interpreting is the 15th fastest expanding job category in the United States, with projected growth of 42% by 2020.

How many times I have heard that a translation is no substitute for the original? Countless. This is a trite remark that is obviously wrong. It’s just a piece of folk wisdom that is just untrue. It’s amazing how many people fall into this trap.

People who say that translations are no substitute for the original imply that they have the means to recognize and appreciate the source text as opposed to its translation. Without this ability they wouldn’t be able to make this claim. If you are unable to tell a Chianti from a Merlot then you can’t possibly compare them. In the same way, the ability to discriminate between a translation and its original implies that you master both languages, the source and the target. How many people have this set of skills? Not too many…

Then the question is: is the average reader able to tell if they are reading a translation? If the translation was made by a professional, absolutely not. As a matter of fact, countless writers in the past have devised originals as translations and vice versa.

For example, in 1761 a minor Scottish poet called James Macpherson claimed to have discovered and translated from the Gaelic an epic on the subject of Fingal, related to the Irish mythological character Fionn mac Cumhaill (Anglicized to Finn McCool) written by Ossian.

The following year the Scottish poet published, to great acclaim, “Fingal, an Ancient Epic Poem in Six Books”.

For many decades, this poem was considered to give precious insight into the ancient culture of the original inhabitants of Ireland. Eminent figures such as the great Napoleon and Thomas Jefferson and also well learned ones like the German philosopher Johann Wolfgang von Goethe and Johann Gottfried von Herder were fascinated by the poetry of Ossian, the “Gaelic Bard”. He was soon proclaimed as the Celtic equivalent of classical writers such as Homer. Many writers were influenced by these works, including one of my all-time favorites: Sir Walter Scott.

But they were all wrong! The story of Ossian hadn’t been invented by Celtic poets at all. It was written directly in English by James Macpherson himlf!

Authors may have different valid reasons for disguising an original work as a translation and vice versa. Sometimes it’s done to get through censorship or to serve individual or collective fantasies about national of linguistic authenticity. What all such deceptions tell us is that by reading alone it’s simply impossible to tell if a work was originally written in that language.

Francesco Pugliano

To me, translation is more an art than a science. And translators are very much like artists. A proof that translation is an art can be found in the fact that what may seem to be a beautifully crafted translation to someone, may not seem so to somebody else.

Jorge Luis Borges

Jorge Luis Borges

Translating is like playing music. Some people may like it, other people may not. Mozart was a great composer, but that doesn’t mean that everybody likes his music! Musicians work with sounds, translators work with words. And exactly like in music sometimes a cover can be better than the original, also translations sometime can be better than the original. The great Argentinean writer Jorge Luis Borges, after reading a great translation once said “The original is unfaithful to the translation.” Meaning that the translation was better.

The role of translators, like that of musicians, is not passive or mechanical. You may know how to play the piano, but that doesn’t mean that you can execute Puccini. At the same time you may speak a second language, but that doesn’t mean you can render the message in the same way it was intended by its original author. Translators need to master not only the language but also what’s behind it.

This is because translators are a sort of a bridge between different cultures, that of the source language and that of the target language. In fact, the word translation comes from the Latin “translatio”, which itself comes from “trans” and “fero”, together meaning “to carry across” or “to bring across”).

I recently found this video on the art of translation. It’s rather long (56 minutes). It’s in Italian with subtitles, but also with some parts in French, Spanish, Portuguese, and German.



The title of the video is Tradurre “Translating” by Pier Paolo Giarolo, published in 2007. The 12 translators appearing in the video are:

Erri De Luca
Fulvio Ferrari
Silvia Pareschi
Luca Scarlini
Nadia Fusini
Donata Feroldi
Elisabetta Bartuli
Rita Desti
Anna Nadotti
Paola Tomasinelli
Maurizia Balmelli
Enrico Ganni.

I find it very inspiring. And I personally have a lot of respect for that translator (Luca Scarlini) at minutes 23, when he says that he started this profession to impress a person who was very important to him. And how he did it? By translating Hugh Selwyn Mauberley by Ezra Pound.

Ezra Pound

Ezra Pound

Here is an excerpt:
Beside this thoroughfare
The sale of half-hose has
Long since superseded the cultivation
Of Pierian roses.

Enjoy the video!

Francesco Pugliano

When designing a mobile app, one of the things a designer should keep in mind is translated text expansion. Probably this doesn’t come as something new; there’s a whole literature on the subject.

However, when designing and localizing a mobile app you should also avoid other layout tricks that as a designer you might be tempted to try. I recently stumbled upon an app for iPhone that struck me for its peculiar use of ellipses.

What are ellipses? As a convention, ellipses in UI text are used in the following cases:

  • To indicate that a command needs additional information.
  • To indicate that text is truncated.
  • To indicate that a task is in progress (for example, “Searching…”)



Normally, when the translated text expands too much, the designer would use an ellipsis at the end of the sentence.

In this case Italian expands by approx 40%. Japanse seems to be aprox the same lenght as the English.

English: New message (11 characters)
Italian: Nuovo messaggio (15 characters)
Japanese: 新しいメッセージ

The funny thing is that in this specific app, the designer used an ellipses in the middle of the text, and not at the end where it should normally be.

As a result, the meaning of the Italian translation is not “New Message” but “New essay”. That’s what “saggio” means in Italian.

Why the developer didn’t stick to widely accepted conventions on the use of ellipses? This is still a mystery to me.

I’ve never been a big fan of ellipses used to indicate that a text is truncated. I believe that a better practice would be to implement the convention of scrolling text. Like a ticker message that scrolls across the screen letting the user read the entire string from beginning to end. On mobiles, this is used a lot in lists in the body of a window. And while I haven’t seen it used in the window title of an app, it would be a worthy experiment for design engineers to figure out.

Food for thoughts. What’s your take?


Introduction to AI

AI (Artificial Intelligence) is the study and design of intelligent agents. AI programs are called Intelligent Agent. Here is how it works:

The Intelligent Agent (on the left) interacts with an Environment (on the right). The Agent perceives the state of the Environment through its sensors and at the same time it affects its state through its actuators.

The real challenge about AI is the function that maps sensors to actuators: that is called Control Policy for the Agent.

Based on the data received from sensors, the agent makes decisions and pass them over to its actuators. These decisions take place several times and the loop of environment, feedback from sensors, agent’s decision and actuators interaction with the environment is called Perception-Action-Cycle.

AI is used in many fields, among which:

  • Finance
  • Robotics
  • Games
  • Medicine
  • And of course: the Web

AI and uncertainty

AI is all about uncertainty management. In other words, we use AI if we want to know what to do when we don’t know what to do. There could be many reasons for uncertainty in a computer program:

  • Sensor limits
  • Adversaries that make it hard for you to understand what’s happening
  • Stochastic environment (where behaviors are intrinsically non-deterministic)
  • Laziness
  • Plain ignorance (many people that don’t know what’s going on, could easily learn it, but they just don’t care)

All of the above are possible causes for uncertainty and AI.

Example of AI in practice

One of the many key applications of AI techniques is Machine Translation. How does Machine Translation work?

Machine Translation generates translations using AI techniques based on bilingual text corpora. Where such corpora are available, impressive results can be achieved translating texts of a very similar kind. Unfortunately, such corpora of bilingual texts are still very rare and the size of the available corpora varies significantly from one language combination to the other.

So what does Machine Translation looks like? On a large scale Machine Translation system, examples are found on the web. On a small scale, they can be found anywhere. This example was found in a Chinese restaurant in Cupertino:

In these type of text a line in Chinese corresponds to a line in English. To learn from this text, we need to find out the correspondence between words in Chinese and words in English. For example, we can highlight the word “wonton” in English. It appears 3 times throughout the text. In each of those lines there is also one Chinese character that appears: 雲. So it seems that there is a high probability that this ideogram in Chinese corresponds to the word “wonton” in English. Please note that we are talking about probabilities here. As a matter of fact “wonton” in Chinese is 雲吞 and not just 雲. For some reason the ideogram 雲吞 on line 65 is abbreviated to just 雲. And it’s not a common abbreviation.

You can go further, and try to find out what ideogram in Chinese correspond to the word “chicken” in English:

Please note that we aren’t 100% sure that 雞 is the ideogram for “chicken” in Chinese but we do know that there is a good chance because each time the word “chicken” appears in English this ideogram appears in Chinese.

Now let’s see if we can find a correspondence for the word Soup:

As you can see the word “soup” occurs in most these phrases but not in all of them. In the English side of the menu is missing in 1 place (65. Egg Drop Wonton Mix). Equivalently, on the Chinese side of the menu is missing in 1 difference place (廣東雲吞 60).

The correspondence doesn’t have to be 100% to tell us that there is still a good chance of a correlation.

In Machine Translation these type of alignment is used to create probability tables. Hence the name Statistical Machine Translation. In other words, the probability of one phrase in one language to correspond to another phrase in another language.

More on Machine Translation in future posts. Stay tuned.

Francesco Pugliano


The end of free Machine Translation API

Last June Adam Feldman (API Product Manager at Google), announced they were pulling the plug on their Google Translate API, causing a lot of concern and some protests in the developers and localization world. You can read the announcement here.

Then in August, Jeff Chin, (International Product Manager at Google) took that back and announced that they were offering the Translate API at a cost instead of free of charge. You can read the announcement here.

Here is Google’s pricing model:
$20 per million characters of text translated.

In September, Vikram Dendi (Director of Product Management at Microsoft), announced something very similar, but not many people took notice. You can read the announcement here.

Here’s Microsoft’s pricing model:
No cost up to 4M characters a month. Then $10 per million characters.

Unlike Google, Microsoft will only charge you when you reach the threshold of 4M characters a month and will then cost half as much ($10 per million characters instead of $20).

Quality of Google and Bing Machine Translation services

The quality of Google and Bing Statistical Machine Translation systems now that the technology is mature, heavily depends on the quality of the parallel text found on the web and crawled by their MT engines. Before the advent of Google and Bing translate, parallel text found on the web – more often than not – was produced by professional translators, and therefore of good quality.

Now, translating content professionally is expensive. Depending on the domain of translation and the language pairs, professional translation can cost as much as $0.50 per word for a language such as Japanese and between $0.18 to $0.21 per word for European languages.

During the recent financial crunch in 2008, many web publishers needed to cut costs. It’s not a surprise that they started to abuse the free Google Translate and Bing Translate API to translate content and then publish it as is, with no professional review.

This is a common technique that SEO companies have been applying to bring more users to a website and then turn them to premium content (professionally translated content).

The problem is that no algorithm is (yet) capable to understand whether content has been translated by a Machine Translation system or by a professional translator. Only trained human translators that speak the language can do that.

Today, both Microsoft and Google Machine Translation engines are crawling and processing web content that may have been published without any human proof-reading after being translated using the very same Google or Microsoft’s translation API.

In other words, these two companies are “polluting their own drinking water”.

I hope that by starting to charge for their Machine Translation Services both Google and Microsoft can decrease or at least control the amount of sub-standard translations published on the web so that in turn their MT engines can produce more reliable translations. Feeding their engine with United Nations and European Union bilingual documents is not enough to produce high quality translation.

Size doesn’t matter without quality

Many publishers in recent years have started to build their own corpora of bilingual texts to feed their Machine Translation engines with. It’s a given that an ad-hoc Machine Translation database fed only with high quality human translated and proof-read bilingual text in a specific domain can produce higher quality than Bing and Google Translate.

Unfortunately at some point these publishers may start to pollute their MT systems with content that has been machine translated and not carefully reviewed by professional translators.

We have seen this happening in the past, for example when the hype was all about Translation Memories instead of MT engines as it appears to be today. Some companies saw their Translation Memories growing bigger and bigger with no or little control on the quality of the content they were fed with, thus polluting their TMs and making them almost unusable.

Francesco Pugliano