Instructions for importing a bibliography into scratchpad using the RIS-format

Written by Stephan Imhof

For those of us who do not use Endnote as the software for reference management, Scratchpads offer three more data formats for importing bibliographic data. For me, the RIS-format turned out to be the less complicated. Many bibliographic software has features to facilitate the export of content via this format.

RIS is a text file which assigns every field of a data set with a preceding tag. This tag always has 6 characters of the following structure: two upper case letters or numbers, two spaces, a dash, one space (e.g. "AU  - "). The letters and numbers specify the data field. Each data field with its tag is to be terminated by a "carriage return/line feed" (hit enter). Each data set must start with the type-tag = "TY  - " (skip the quotes) and must end with an "ER  - " (probably means ‘end of reference', does not have content). The other tags can be in any sequence. The meaning of the tags can be looked up on the following website, which in any case is most instructive to learn about RIS: http://www.refman.com/support/risformat_intro.asp

So far everything sounds quite easy. However, the import into Scratchpads is strongly dependent on the first tag, the "Type" of the data set. If this type is set to be BOOK it will ignore some fields, e.g. the "A2  - " or "ED  - " for editor, and if it is set to JOUR = journal article, it will handle the A2-tag like a ‘secondary author' and not as an editor, and it will skip anything which is tagged with "PB  - " for publisher and "CY  - " for the publishers city. For my purposes the division into three ‘types' turned out to be sufficient, so far. BOOK for those references which are whole books, CHAP (= book chapter) which include editors (will take publishers data, too), and JOUR for journal articles. In my bibliographic database the ‘type' was not a standard data field, thus, I had to set up a new data field and fill it with the three words mentioned above for an automated import.

Your text file exported from your source database should look something like that:

TY  - CHAP
T1  - Seed structure in Voyria primuloides Baker
(Gentianaceae): Taxonomic and ecological implications
A1  - Bouman, F.
A1  - Louis, A.
Y1  - 1989///
SP  - 261
EP  - 270
CY  - Amiens
PB  - Univ. Press. Picardie
A2  - Paré, J.
A2  - Bugnicourt, M.
A2  - Mortier, J.
A2  - Juguet, M.
A2  - Vignon, F.
A2  - Vignon, J.
T2  - Some Aspects and Actual Orientations in Plant Embryology
ER  -

TY  - BOOK
T1  - Voyria
A1  - Aublet, J. B.
Y1  - 1775///
SP  - 208
EP  - 211
VL  - 1
CY  - London, Paris
PB  - P.-F. Didot jeune
T2  - Histoire des plantes de la Guiane francoise
ER  -

TY  - JOUR
T1  - Characters of a new genus consisting of two species of parasitic Gentianeae
A1  - Gray, A.
Y1  - 1871///
SP  - 22
EP  - 23
JF  - J. Linn. Soc., Bot.
VL  - 11
ER  -

You must ensure to have the correct character encoding, which is ‘Unicode UTF8'. This is particularly important for those using special characters like ü's, ä's and ß's. If the encoding is wrong every word which includes a special character will be truncated starting from this character. Irina Brake pointed me to a little software called ‘EditPad Lite', which can convert e.g. western European character set to UTF8. EditPad Lite is freely available at http://www.editpadpro.com/editpadlite.html. Open your file in EditPad Lite, go to 'Convert', check 'Encode the original data with another character set...', select 'Unicode, UTF-8, check the Preview, click on 'OK' and save your file. Importing the converted text file results in correct appearance of the special characters.

Now you are ready to go to the import page of the bibliography module in your Scratchpad. I strongly recommend to test the import in the sandbox first (http://sandbox.scratchpads.eu/). And, initially pick only a few characteristic data sets for a test. The system does allow the deletion of wrongly imported references, but only 50 at once (if I remember right). This means, if you import 800 references, and they are all wrong, you need to delete them in 16 steps (takes you about quarter of an hour).

The import form is self explaining, choose your file, choose RIS as the format and click ‘import'. Next you can go to ‘bibliography' and admire your work.

Don't be surprised when some of the content in your data sets does not show up in the regular Scratchpad bibliography view. This view, as I understand it, is just a teaser, not showing all data belonging to this data set. Click on the title of the reference and you will see the missing information.

If you add references, be aware of the fact that Scratchpads do not check for duplicates. In order to avoid double or triple entries it might be a good idea to mark the references in your source database which you already have exported.