October 14, 2015

Summer School Experience: TEI: What? How? Why?

TEI: What? How? Why?

So what is TEI? TEI stands for the Text Encoding Initiative and is a consensus-based means of organising and structuring your (humanities) data for long-term digital preservation. It is well-understood, widely used, and prepares your data for presentation in a variety of digital formats. Originally developed in conjunction with manuscripts but flexible in application it can be used as a means of creating meta-data for an item, ie. title, author, provenance etc., or for transcribing the item itself. The excellent talk I heard at the DHOxSS on this topic was by James Cummings (ITS, Univ. Ox.).

You can tag elements such as <HI>this</HI> to control the presentation of your document on screen, or names such as <name>James Cummings</name> in order to make the document searchable. Another benefit is that compared to other computer languages it is easy to read as a human, and thus easy to work with if it is your first time behind the scenes of a ready-made document. The TEI handbook weighs more than a conference lunch, but one of the most attractive aspects of the language is that if when describing your text you find that what you want to describe isn’t in the handbook (which when you consider how fussy academics are is quite likely, and furthermore is why the handbook is so thick to begin with) you can simply invent your own tags:

<p><l><noun clause><definite article><capital>T</capital>he</definite article> <adjective>key</adjective> <noun>thing</noun></noun clause> <verb clause><unidentified grammatical element>to</unidentified grammatical element> <verb>remember</verb></verb clause><punctuation>.</punctuation></l></p> … is that you don’t have to tag everything! Creating a TEI document is not an end in itself, tempting as this may be to the more <gap/> amongst us. It is a means to preserve and describe your document in order to accomplish specific predefined ends. Are you interested in social awareness of the new world? If so, tag places. Are you interested in the regional creation of manuscripts? If so, tag local spellings. A recent project which has made extensive use of TEI in my field of early modern drama is Richard Brome Online as discussed by Eleanor Lowe (English and Modern Languages, Oxford Brookes) at the DHOxSS. In this resource there are particular tags for speakers, stage directions, proprs, or acts and scenes (TEI privileges sense over physical arrangement of the text in the source, ie. paragraphs over pages). These provide information both for displaying the text online, and making it searchable by the user.

If you’re not creating an online edition, which most of us currently aren’t, it is still helpful to understand TEI when one encounters these texts, as it is useful to understand what choices went in to the creation of the text you’re reading. It is furthermore useful to have some working knowledge of TEI for the small scale preservation of original manuscripts. It allows for accurate representation of the document for perusal after leaving the rare books room. The focus of your transcription might be on text or physical pre<stain “type=coffee”>sentation,</stain>but in either case the key to success in using TEI is consistency.

Clearly the tags I have been using are not necessarily ideal for all projects. Comprehensive guidelines are available at webpage (above). The ElEPHãT project, as discussed by Pip Willcox (Bodleian Libraries, Univ. Ox.), has with the Text Creation Partnership combined data from HathiTrust, ESTC (English Short Title Catalogue) and EEBO (Early English Books Online) in an extensive TEI project. Various transcriptions of the texts that they have worked on are available online at EEBO. Their work can be utilised via basic searching in order to find places or people of interest to your work, or developed by the user in order to conduct more particular research of their own devising. In a workshop session where we were left to play with the TEI documents available from the TCP I suggested that one could devise numerical tagging across their strong collection of alchemical texts in order to investigate the prevalence of sacred numerology in these works. Unfortunately this may have to wait for another day.

In conclusion:


  • TEI is a language which can be used to preserve the data and metadata of early manuscripts and texts.


  • The text is inputted manually or, if you’re lucky, via OCR, and marked up (tagged) according to a pre-considered range of purposes to which the data is likely to be put.


  • TEI separates the data from the interface and so ensures that the hard work put into digitising the documents won’t be made redundant once the interface becomes obsolete.
  • It is flexible and so can be tailored to the needs of any project.
  • It is widely understood allowing for sharing of data.
  • It can be read by both computers and humans and so is relatively easy to get to grips with.
  • It can be used for the smallest and largest of projects.
  • Trust me, its fun!

Emil Rybczak (English) Univ. Warwick.

- No comments Not publicly viewable

Add a comment

You are not allowed to comment on this entry as it has restricted commenting permissions.

October 2015

Mo Tu We Th Fr Sa Su
Sep |  Today  | Nov
         1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31   

Search this blog



Most recent comments

  • Thanks for this short blog by Dave on this entry

Blog archive

Not signed in
Sign in

Powered by BlogBuilder