[Tshwanelex-l] Building a Name Dictionary (Trauring)

Geoff TOISTER gat178 at 012.net.il
Fri Apr 26 05:05:04 EDT 2013

Hi Philip,

Before I give my perspective on your specific questions, a piece of general advice is to print out the User Guide and study it carefully. Many of your questions are answered by the guide, as well as by the YouTube demos and the online FAQs and examples.

You would have to ask TLex how long the trial version works for and what features are missing.

I have not used TLex for building a name dictionary. I am currently using tlCorpus for a schools’ dictionary project, and it does seem to me that it could be used to build a name corpus.

The Specific Questions

1. The parts of speech are editable, and this is covered on p.61. “Attribute Lists”. Thus, you could rename the POS attribute and call it “Language of Origin” and the list could include options such as English, Greek, French, etc instead of noun, verb, adjective, etc. You could also include the original spelling of the name – in any script you choose (e.g. Greek, Hebrew) –  by utilizing the “Translation Equivalent” (TE) facility (see p.61.). The TE can be positioned anywhere within the entry. Thus, it could be formatted so as to appear to be a run-on of the “Language of Origin” attribute. Alternatively, you could utilize the Lua Script “smart styles” facility (p.84). I have not tried this myself, but my understanding is that it could generate attributes formatted like (Greek, Philippos), as in your question.

2. I am not conversant with InDesign. However, I do know that whatever formatting is set in the DTD is fully preserved when exporting. This includes colours, fonts, spacing, etc. etc. I also know that the TLex exporter offers the option of exporting in columns (at least two – I’m not sure about ‘multiple’) and there is a guide to exporting on pp. 91-95 of the user guide. I don’t understand the question about ‘links’ but if you get an answer to that question from someone else, I would be grateful for a copy. I am also interested in creating interactivity on a digital platform.

3. I don’t think there is any way to dynamically link external data to TLex data. But a non-dynamic solution is available by way of the frequency facility (p.34). You could easily import the rank numbers from your spreadsheet and have them displayed as an attribute (p.61.).

4. TLex provides a default DTD which can be modified for the specific purposes of any dictionary (e.g. a name dictionary). Various template DTD’s are also available. Of course, you could also set up a unique DTD from scratch. Easy or difficult is a question of how much time you have for the job. I would suggest you set up a pilot project for approximately 100 lemmas (i.e. names) and develop a new DVD by means of trial-and-error testing. A Lua filter (see p.103) may provide a means of setting up a section of the entry that is dependent on something being true. I’m not sure on this, not having tried it myself. TLex never wastes space. If any one element is defined in the DTD, but not populated in a particular entry, then the next populated element follows on immediately from the previous one. As for notifying you whether or not all entries marked as biblical had associated passages assigned, this can be easily done – during the compilation stage – by tagging and filtering. A thorough demonstration of this technique is available on YouTube.

5. In one word, the answer is ‘Yes’. Read up on elements and their attributes on pp. 56-57. The relevant attribute in this case is ‘etymology’, which is available by default in the standard DTD. I cannot give you an authorative answer about linking an index to a page number, but I imagine that it could be done by someone who has programming skills in both TLex and InDesign. Once again, if you get an answer to that question from someone else, I would be grateful for a copy.

6. XSL or CSV data can be imported into TLex. However, the success of the procedure depends on strictly mapping the column headings to the relevant TLex element. This would mean that you would have to standardise the format of all your spreadsheets first, set up your ideal DTD, and then do some trial runs to see if the import presents the data satisfactorily. However, as I understand the question, your data is not actually a corpus in the same sense as the word is used in lexicography and by tlCorpus. If you do want to go that way, and if your original sources are in .doc files or .txt files, you could use them to build a unified tlCorpus corpus (the process is quick and painless) and to use it instead of the Excel files. This would enable you to view all the instances of a particular name, to get a frequency count on each name, to view the name in context (i.e. with X words before and after), to view the entire original text and to copy-paste chunks of text from tlCorpus to tLex.

I hope you will find this information helpful.

All the best
Geoff Toister
This message has been scanned by AVG anti-virus

No virus found in this message.
Checked by AVG - www.avg.com
Version: 2013.0.2904 / Virus Database: 3162/6272 - Release Date: 04/25/13
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.yale.edu/pipermail/tshwanelex-l/attachments/20130426/4fc6926b/attachment.html 

More information about the Tshwanelex-l mailing list