Re: transcription from handwritten text
Christian Thomas <[email protected]>
| Newsgroups | gmane.text.tei.general |
|---|---|
| Message-ID | <[email protected]> |
Hi Paul, Am 27.01.2017 um 21:47 schrieb Paul Schaffner: > Has anyone used an outside firm or agency to transcribe and apply > markup to modern handwritten text? Or can recommend any capable > of doing so accurately? thanks for your question, like you, I would be very interested if other people would have tried that,! I had the expression it was not very common, but should be considered as an option more often. Anyway, I can really recommend it. In the Hidden Kosmos project (http://www.culture.hu-berlin.de/hidden-kosmos) we worked together with a vendor, textloop (cf. http://textloop.de/ and http://textloop.de/kosmos/), to transcribe 19th cent. handwritten text. The vendor transcribed and annotated a total of ca. 2500 mansucript pages in record time and with outstanding accuracy. They delivered full-blown TEI accordung to specifications developed based on the DTA Base Format (DTABf, http://www.deutschestextarchiv.de/doku/basisformat) with some additional tags specific to handwritten texts like additions, deletions, etc. Susanne Haaf from the DTA presented some of these aspects at the TEI-MM 2015 in Lyon, and there is an upcoming paper on the DTABf for manuscripts to appear in the jTEI. You can see the results of textloop's work here in the Deutsches Textarchiv collection, and that way you also get an impression of the difficulty of the manuscripts. These are different student's notebooks documenting Alexander von Humboldts famous Kosmos-Lectures (1827/28). textloop transcribed these five notebooks: http://www.deutschestextarchiv.de/patzig_msgermfol841842_1828 http://www.deutschestextarchiv.de/libelt_hs6623ii_1828 http://www.deutschestextarchiv.de/nn_oktavgfeo79_1828 http://www.deutschestextarchiv.de/nn_n0171w1_1828 http://www.deutschestextarchiv.de/parthey_msgermqu1711_1828 It was ideal for us to work with a vendor for this project since we could focus on meticulous quality assurance, deeper annotation, cross-linking the documents, identifying scribes, tagging <persName>s and many other things that we would never have gotten round to if we would have had to transcribe the manuscripts ourselves. I am sure Martina Goedel from textloop will be happy to tell you more about how she organised her work and what this experience was like for her -- since transcribing manuscripts (and not prints) via vendor still seems to be unusual and, in the Hidden Kosmos project, was a 'first' for both sides. Again, I'd be very interested to hear if others have experience with that workflow. Also, Paul, I'd be interested to see what the manuscripts you consider to have keyed by a vendor look like and, of course, am curious if this works out for you. Best wishes Christian Christian Thomas Hidden Kosmos: Reconstructing A. v. Humboldt’s »Kosmos-Lectures« http://www.culture.hu-berlin.de/hidden-kosmos Humboldt-Universität zu Berlin Kultur-, Sozial- und Bildungswissenschaftliche Fakultät Institut für Kulturwissenschaft Mohrenstraße 40, 10117 Berlin Raum: 414, Tel. +49 (0)30 2093 66 147 E-Mail: [email protected] -- > > (This is a bit off-topic, so you may want to reply off-list.) > > Paul