Glosbe provides an API to access dictionary data in multiple languages. It's easy enough to grab the html or JSON, but putting it in a "pretty" format is the next challenge.

Also on the to-do list. Allow multiple calls to the API in order to fill in the blanks of a list of words created during the course of personal reading or movie watching. This way I can keep track of the words I am learning and not have to do the footwork of looking for the word.

A central place to store the list is also something I would like to experiment with. Evernote is the first application, crossing over the mobile-desktop boundary seamlessly.

1) Keep a separate database of your translation memories bundled together into one .db file. Use APSIC Xbench or TmLookup to perform searches to complement the specific project translation memory you have open in your working CAT tool. You can use Xbench for many file types and you can count on it to be compatible with almost any file type since Xbench is a true multi-tool. When you're in a bind and need to convert from one file format to another, look no further. See this tutorial on how to create tmx and txt files out of your native TM file.

2) Have a browser open to your most used search engines and online databases. For me these would be the Real Academia Española dictionary page, a Proz.com glossary page where I can look at discussions on terms other translators have seen before. Go to proz.com/search. For anything that sounds remotely like legalese or economics jargon, access the EU DGT website.. IATE and EUR-Lex have other important official language databases. Some others are listed by a prolific purveyor of translator tips, Pablo M. Sánchez: 

3) Get access to a standard industry database or body of knowledge that can provide you with concepts that should/must be reproduced with a certain wording in your target language. You may even be able to get a quick look at what types of concepts a document contains without reading it. METAMAP from the National Library of Medicine (NLM) is one such tool that can process medical text (so far only in English, but developers are working on extending the functionality to other foreign languages) to output known medical concepts it found in the document using its ontology of medical concepts. To access the NLM tools, you need to apply for a license.


4) MultiFultor, by Rolf Keller, allows you to conduct concurrent searches in your favorite online or offline resources without having to touch your browser. The documentation is around 30 pages. After testing it out, adding websites and using the tool to ensure access to my sites, I am able to say that this is a tool that is easy to set up. and perfect if you need to conduct fast internet searches but do not want to leave the translation alone for very long, whether you are in the zone, on a tight deadline, or both.


5) There are some great Google search operators to use if you really want to drill down and find specific types of documents on subjects you are looking for, e.g. the words "vascular" and "drug" in the title index from an actual document like a pdf or .doc file.
- intitle:(vascular + drug) filetype:(pdf or doc)


Most internet resources don't offer the assurance of quality that your trusted print resources do. Don't rush! Read the sentences before and after the concept you want to learn more about. (Context is everything.) Then start your vocabulary research with monolingual and bilingual dictionaries.. I recommend keeping general and specialized versions on your desk and referring to them in conjunction with more specialized resources, e.g. Black's Legal Dictionary or similar, to get a well-rounded vision of the term you are studying. 

If you have any favorite practices you would like to share, please leave a message below!