Preparing your laptop for a TXM workshop

The objectives of the TXM Initiation workshop are not only to teach you how to use TXM, but also to allow you to do it in your own working environment, and to leave the workshop with TXM installed and configured properly (the configuration of some parameters, such as results export, is finalized during the workshop).

That is why we recommend everyone:

  • to come with their own laptop, and
  • to install TXM (and the TreeTagger associated software) before coming, to avoid taking on the collective time of the session.

Installation is a priori straightforward, but you know that with computers you can always be surprised, hence our caution.

Here is some useful information for installation.

General note for downloads

The TXM software is hosted for free on the SourceForge site. When you click on a link for downloading TXM or a related resource (manual, corpus…), a new page is displayed, in which:

  • at top left you see progress information indicating that the download starts, with a five-second delay: just wait before the usual procedures for a file download start automatically;
  • on the largest part (middle, margins) very conspicuous links that may also be called “download” can appear, these are advertisement links that you should ignore.

Note 1: if you want to avoid advertising links on the web in general, we recommend:

  1. and to install the Adblock Plus plugin in this browser: https://addons.mozilla.org/en/firefox/addon/adblock-plus

Note 2: We are interested in hearing about hosting services equivalent to Sourceforge and which would prevent such advertisements.

TXM download and install

The current version of TXM is 0.7.6. Reference instructions for installation are given in the readme associated with this release, text that appears below the list of files available for download:
http://sourceforge.net/projects/txm/files/software/TXM/0.7.6

For Mac users (especially for the 10.9 version): Beware there is a known bug in some Macs with the 0.7.6 version of TXM, see the description and what to do if you are in that case in the txm-users mailing list message of December 5 2014: https://groupes.renater.fr/sympa/arc/txm-users/2014-12/msg00001.html (in French)

In case of additional information need or installation difficulty, please first consult the FAQ (Frequently Asked Questions) section of the users wiki (in French):
https://groupes.renater.fr/wiki/txm-users/public/faq

For example, if you are not sure of the architecture to choose for your machine (32-bit or 64-bit?), this section of the FAQ can help you: Dois-je installer un TXM 32 bits ou 64 bits ? (Do I need to install a 32-bit or 64-bit TXM?)

A common problem is to have the right Java installed, you can find help here: Au premier lancement TXM, ne démarre pas et j'ai un message d'erreur comme quoi TXM ne trouve pas Java, pourtant je l'ai installé ? (At TXM first launch, it doesn't start and I get an error message saying that TXM doesn't find Java, yet I have installed it?)

The FAQ is constantly evolving so remember to check the latest version when you need it.

Once the installation process is complete, you can make an initial check on the proper installation of TXM:
comment_verifier_que_txm_s_est_installe_correctement (How to verify that TXM is properly installed?)

Furthermore, the TXM Manual is available online:

  • in PDF version for download (in French)
  • in HTML version for online consultation (in French)

There exists an English translation of the TXM 0.5 Manual. You can use it for general introduction and for basic search and analysis functions, however the interface has changed significantly since the TXM 0.5 release.

We will give you a printed version of the manual at the beginning of the workshop (for workshops that take place in Lyon).

TreeTagger download and install

Installation and setting of the TreeTagger associated software (for on the fly morphosyntactic tagging and lemmatization of the corpus):
http://txm.sourceforge.net/installtreetagger_en.html (the French version of the page may be more up to date)

If possible, install the English language model (in order to process the text samples used in the workshop). Furthermore it is of course interesting to also install language models corresponding to the corpora you plan to analyze with TXM.

You can then test that TreeTagger is operational in your TXM:
comment_verifier_que_treetagger_est_bien_parametre_dans_txm (how to verify that my TreeTagger is properly set in TXM?)

In case of problems, you can find help in the FAQ:
treetagger_ne_fonctionne_pas_comment_bien_regler_treetagger_pour_txm (My TreeTagger doesn't work. How to properly set TreeTagger for TXM?)

LibreOffice

The results tables exported by TXM -seen in the initiation workshop-, as the texts metadata tables imported with your corpora -seen in the corpus preparation and import workshop-, use the .csv file format. For handling CSV files, it is recommended to have LibreOffice or OpenOffice installed on his computer:
https://libreoffice.org

It is possible to use another spreadsheet program such as MS Excel, but it is much more complicated (Excel “hides” a lot of parameters to simplify common usage) and we usually do not have time during the workshop to resolve all the complications that can arise for different environments. Nevertheless, you can find help on the FAQ for some questions, for example:
Comment ouvrir dans un tableur l'export d'un tableau de résultats de TXM ? (How to open in a spreadsheet an array of results exported from TXM?)

Working corpus for the workshop

Common Corpus (essential)

Just download the file and save it where you want on your hard drive to have it available at the time of the workshop. We will load it into TXM together.

A document that you choose for rapid import into TXM (optional)

During the Initiation workshop, we will experiment the simplest import, called “clipboard”, based on a simple copy&paste.

It is an opportunity to see what happens on any of your files. You can just think before you come to put in your computer a file representing a text (in the broadest sense: it can be a record transcription for example) of a dozen pages (let's say between three and a hundred!), or a small corpus (or a sample of a large corpus) set as a single file, rather in a language for which you have installed a TreeTagger language model (parameter file). Take a format for which the “copy” command works: a file that you open in your word processor (.doc, .docx, .rtf, .odt, .txt…), a web page (rather long); avoid .pdf format or check before that the text content can be selected and copied (no “image” pdf).

Help

If you have trouble installing TXM, and you have consulted the FAQ (for all questions concerning installation and launch), and your professional colleagues or staff can not help you, contact us (textometrie AT groupes.renater.fr).

public/preparation_ordinateur_en.txt · Dernière modification: 2017/10/20 15:14 par slh@ens-lyon.fr