Present translation memories store pairs of sentences. Kim’s translation memory would keep that, but would also keep pairs of documents with all of the pictures and formatting. PDF format would be good, see synchronising PDF documents too. This would mean keeping far more context than is currently the case, for almost zero effort. All you need is a bit of online storage and a fast connection.

Today i came across Schwerpunktstation. Both Leo and IATE call this a “distribution network core unit”. I have never heard of that, so I did the Google test and got just 1 hit from a Dutch website. So, it’s obviously not called that in the Anglo-Saxon world. I think that I will choose the closest quickly understandable term and call it a substation. Wouldn’t it be nice though if online dictionaries could do a plausibility check like this, as well as using votes from users, in a similar way to Websters Online “usage frequency” ?

I am often frustrated by abbreviations in German. I suppose they are so popular as some expressions are so long in German. For example, in Berlin, everyone knows that BVG means Berliner Verkehrsbetriebe, ehemals Berliner Verkehrs (Aktien)Gesellschaft. When you have just arrived from Birmingham, it takes a while to work out what they are talking about.

What I would like is a tool that does a German spell check in a single pass, requiring no manual intervention, just producing an alphabetically sorted list of the failures. Those words that fail the spell check would include most of the abbreviations and give me a list to send to the customer at the start of the job. There is then a reasonable chance that answers would arrive before I have finished.

Of course I could just collect them manually as I work though, but that means I can’t send the list until I have finished. When I have finished, I don’t really want to go back a few days later, when the answers arrive, and open up the job again, when I am in the middle of something else.

This is a feature I would like to see included in translation tools

Context
Context is vital to produce the correct translation. Most translation tools however remove all of the context and offer a confusing jumble of tags and words from just a single sentence, the one that you are currently translating. Any pictures are stripped out, the position and role of the sentence on the page is lost as is the presentation which was carefully crafted to help you to understand the content.

The cause of the problem
This arises from the use of differing editing software to produce the original documents and the need for conversion tools to take a variety of source formats and produce one consistent format which can be understood by the translation tool. When the translation is complete, the inverse conversion is done to generate a document in the original format, ready to be opened by the original editing software.

The solution
My workaround to this is to open up a copy of the source document in PDF format. Whatever the original editing software was, if it was a document intended for printing, a PDF version can be produced at the click of a button. I copy a tag-free part of the sentence to be translated from the translation tool and search for this in the PDF document. Obviously I can’t do this for every sentence as it would take far too long.

If I can do this manually, software can do it too, and that is the feature that I would like to see in all translation tools.