About this blog

Translator's Shack is a collection of links, news, reviews and opinions about translation technologies. It's edited and updated by Roberto Savelli, an English to Italian translator, project manager and company owner of Albatros Soluzioni Linguistiche, a team of English-Italian translators, which hosts and supports this blog.

The Life as a PM category, managed by Gabriella Ascari, contains topics that are less technical in nature, but which we're sure will be appreciated by owners of small translation businesses and freelancers.

Here are links to my pages on some social networks:

Highly recommended:

Jost Zetzsche video – Reconvergence of MT and TM

The ATA Language Technology Division page contains an interesting video featuring  Jost Zetzsche. In the video, Jost explains how machine translation has rapidly evolved from a separate, quite isolated technology into a new concept that is very much integrated in other translation tools and systems used by human translators.

Jost goes on explaining the three main supporting arguments to his theory:

  1. SDL, among other providers, has integrated MT into its mainstream translation memory tools. This means that translators are able to leverage suggestions from SDL’s own generic MT engine (which, according to Jost, very often produces a lot of “garbage”). Translators working for enterprise clients that have an account with SDL will also benefit from the client’s customized MT database, which supposedly offers better quality.  It is expected that all the major CAT tool providers will soon follow suit.
  2. Google will soon introduce its Google Translation Center, which will allow translation buyers and providers to use a common platform for the exchange of translation jobs. Such platform will heavily rely on MT and TM technologies.
  3. The initiatives of  TAUS – Translation Automation User Society are aimed at pooling the translated material of very large translations buyers (among which the EU, Microsoft, Oracle, etc.) in order to obtain better results with machine translation by leveraging the enormous amount of translated material produced by these organizations.

The conclusion? According to Jost, translators will not be able to oppose the radical change that all this will cause, so they’d better face the music and start learning new skills, such as machine translation post-editing.

Interesting article on Google’s machine translation system published on “Thoughts on Translation” blog

Corinne McKay’s Thoughts on Translation blog contains a very interesting new post and comments about Google’s machine translation system. She experimented with the system and reported her findings.

I’ll go on the record as saying that I think that most translators are much too paranoid about machine translation. MT technology has come a long way, and I think that we’re on the cusp of its being considered a standard productivity tool for translators, much as translation memory is today. However, I think it’s not worth losing sleep about machine translation sending human translators the way of the telegraph operator.[…]

So, I decided to do my own unscientific test of what Google Translate is producing these days, using three different texts that are similar to what I might be translating on any given day: fluffy, dense and flowery. Let’s see how it did.

via Thoughts on machine translation « Thoughts On Translation.

Compendium of translation software: directory of commercial MT and

The Compendium is a long list of  translation environment and machine translation tools, sorted by product name and by supplier, and containing links to the relevant websites. At 100+ pages, this PDF file is a very exhaustive source of information for people interested in computer-assisted and machine translation. New entries are in green.  

Compendium of Translation Software

directory of commercial machine translation systems and computer-aided translation support tools

compiled by John Hutchins

on behalf of the European Association for Machine Translation and the International Association for Machine Translation

Current edition (15th edition, January 2009) [PDF, 605KB]

via Compendium of translation software: directory of commercial MT and translation tools.

Google Translate now automatically detects the source language

Google Translate can now detect the language of the source text pasted by the user. This service is rapidly growing into a solid option for fast , casual translations where the user needs to grasp the general meaning of a foreign-language text. In general, the translations seem to be better than the ones offered by other machine translation programs and services.

Here are some highlights about the service:

  • 22 languages with bi-directional translation in all combinations, for a total of 506 language pairs
  • Auto-detection of the pasted source text (first option in the source language drop-down button)
  • Google Translate’s APIs allow developers to connect their applications to the service. It will be interesting to see which one of the many TEnT providers first comes up with a good implementation of this service. For instance, when a sentence has no matches in a translation memory, the TEnT could quickly query Google Translation and offer a draft automated translation that the human translator can adapt and proofread

Google Translate [via Google Operating System]


This software is a free multi-lingual aggregator of several on-line machine translation services, among which translate.google.com and Systran. The download and installation are pretty straightforward, but in our experience the interface was not very intuitive and the check boxes tended to be rather unresponsive. However, the software works well (if your expectations of machine translation aren’t too high) and fast. Here are some screenshots:

Paste the source text into the bottom pane. The “Services” tab contains the translation engines that are available, based on your language combination and topic fields (see further below).

The “Languages” tab contains all the language combinations available. The availability of any given combination, however, depends on the choices made in the next tab, “Subjects”. Not all translation engines are available for all languages, so your best option may be to select the subject first and then check if your language combination is available in the “Languages” tab.

The “Subjects” tab contains specialization fields that you can choose based on the type of source text you need to translate. Unfortunately, “Law” seems to be available for Italian, but if you select it the language combination becomes unavailable, so we had to stick to “+Common”.

After you’ve figured out your best options based on the language combination and subjects, click on the “Translate” icon on the bottom left corner. The top pane will contain the translated text delivered by the available translation engines. You may get error messages, for instance if a service is currently unavailable or if your source text is too long. In the screenshot below you can see the EN->IT translation of the Apache 2.0 license from the Google engine.

The excellent freewaregenius offers an in-depth review with descriptions of the program’s features.

Translate.Net home page