Hallo!
I have arrived at a point where I have a model which works using a convetional text editor.
It is supported by three CSV files:
Sentences being translated are marked with number of words as are any phrases. [13].
1. English German dictionary, a self-buildt object. Built form my own desktop word usage.
2. An English german phrase glossary of several thousand entries, also from the desktop usage and
3. A growing collection of translated sentence pairs.
For these three I would like to have a customisable interface, such that I can.
1. For the dictionary post a phrase for look-up with head-word marked to direct the look up on three words (typically)
2. For the glossary find all entries with a particular string (the lookup should be 100% hit) and
3. For the sentence pairs (the lookup could be from 70% hit to 100% hit).
Does anyone know of such a script?
How difficult is it to make from a layout definition?
I have downloaded LineByter to test it.
Kind regards, forkinpm.

Hallo Jody!
Over the weekend I had time to reflect on the material I had sent you & have decided to make some changes.
The changes will cover:
1. Being able to handle dual words and a single part-of-speech as in verbs taking a preposition
2. Conjuctions made up of more than one word and also taking a single part-of-speech
For these two points some preview and marking of the text before tagging is required.
3. Some analysis of my texts and their segmentation looking for patterns.
When I have completed this, I will better understand the implications.
This latter point is a way to predict with some better accuracy where phrases begin and end.
The final point is the addition of a rules set for the segmentation and its re-sequencing for the German version.
I have set myself a deadline of mid-day tomorrow to post it.
How are you getting on with what I sent on Saturday?
Regards, forkinpm.