LEMDO has set up a number of diagnostic checks for your edition that you will find useful as you prepare your edition. These diagnostics provide statistics about characters, entrances and exits, and spoken lines. Diagnostics also help you to encode links correctly to anchors from your collation and annotations. For that reason, we recommend that editors regularly check their edition diagnostics. This documentation will guide you through the process of checking your edition diagnostics and clearing any potential errors therein.

To run edition diagnostics, first open the file for your modernized text. In the tool bar at the top of your Oxygen window, click the red play button:

It typically takes some time to run the diagnostics. Once they are complete, a new tab will automatically open in your default Web browser with an HTML page listing diagnostics. At the top of the Web page will be the date that the diagnostics were generated and the ID for your edition. Each set of diagnostics and statistics is available through a drop-down on the HTML page.

Note that diagnostics that have no potential issues will have the number zero (0) beside them, while those diagnostics that do find potential issues will give the number of issues you need to resolve.

The first drop-down on your edition diagnostics page is for statistics. This section includes counts for number of acts, scenes, speeches, stage directions, and speaking characters in your edition. In addition, the length of your modernized text is given in both characters and words. Finally, you can see the total number of elements and attributes that you and the LEMDO team have added to your file.

The statistics section also offers statistics by character. In this table, each character in your edition is listed along with the number of speeches they give, number of words that they speak, their average speech length, the average length of words that they speak, the number distinct words that they use, and how many of their total words are distinct (given as a decimal between 0.0 and 1.0 with 1.0 meaning each word spoken is distinct). This information may be helpful to you when you write your critical introduction.

LEMDO uses anchors and pointers to link annotations and collations to modernized texts. All <anchor> elements should be linked to from somewhere in your edition. This diagnostic finds <anchor> elements that are not linked to from anywhere within the edition.

To resolve this diagnostic, you will first check that the anchor is not being linked to from another edition by searching the texts directory in the LEMDO repo. (Other editions should not be linking to anchors in your edition. They should be linking to the acts/scenes and speeches in your modern text, speeches or WLNs in your semi-diplomatic text, and <div> s or paragraphs in your critical paratexts.) Follow these steps to check that no other edition is linking to your anchors: If there are no files linking to the anchor (as is likely the case, given LEMDO’s general diagnostic that looks for such links), only one result will come up (the anchor itself). In those cases, you may delete the unused anchor. If there is another edition linking to the anchor, leave the anchor in place but let the LEMDO team know that someone is linking to your edition so that we can help the other editor link in the LEMDO-allowed ways.

All pointers must link to an entity with an xml:id. This diagnostic finds <ptr> elements that are trying to link to non-existent xml:ids. If a file is committed with a pointer not pointing at anything, the LEMDO build will break. Checking this diagnostic before you commit files containing new <ptr> elements prevents the build breaking on this issue.

To resolve this diagnostic, search your edition for the pointer that is not pointing at anything and correct the value of its @target attribute so that it correctly links to an entity. These errors are almost always typos (“3411” when you mean “341”.

When annotations and collation link to anchors in a modernized text, the first target must link to the <anchor> element that comes first in the modernized text while the second target must link to the <anchor> that comes second. If the anchors are invoked in the wrong order, our processor will not be able to mark the span and the LEMDO build will break. Checking this diagnostic before you commit your annotations or collation file prevents the build breaking on this issue.

To resolve this diagnostic, search for the annotation or collation entry that links to anchors in the wrong order and correct the order. Run your edition diagnostics again to ensure that the issue is fixed.

LEMDO allows you to make links from <pb> (page beginning) elements to facsimile images using the @facs attribute. Because all facsimile images are associated with xml:ids that are consecutively numbered and we link to facsimile images for all pages in semi-diplomatic transcriptions (including blank pages), we expect the links to follow the pattern of consecutively numbered xml:ids. This diagnostic finds any inconsistencies in the expected pattern.

To resolve this diagnostic, check that your @facs links are correct. See Link to Facsimiles from Semi-Diplomatic Transcriptions for more information.

All speeches in modernized texts should have speech prefixes encoded using the <speaker> element. This diagnostic finds <sp> elements (speeches) without a child <speaker> element. If you have a compelling reason not to give a speech prefix to a speech, take up the matter with your anthology lead(s), who will in turn take the matter to the LEMDO team.

To resolve this diagnostic, open your modernized text file and add <speaker> elements to speeches that do not yet have them. For more information on speech prefixes in modernized texts, see Encode Speakers in Modernized Texts.

Note that not all speeches in semi-diplomatic transcriptions will have speech prefixes. See Encode Speech Prefixes in Semi-Diplomatic Transcriptions and Speeches without Speech Prefixes for our encoding practice in semi-diplomatic transcriptions.

LEMDO lists all of the entrances and exits encoded in each scene of your modernized texts in the entrances and exits section of the edition diagnostics page. This information is often helpful to editors determining scene divisions. It is also useful information if you want to create a doubling chart.

If the list seems to be missing an entrance or exit, check that your encoding is correct. All entrances and exits should be encoded using the <stage> element. For entrances, ensure that the @type attribute on the <stage> element has the "entrance" value on it. For exits, ensure that the @type attribute has the "exit" value on it.

For more detailed information on encoding stage directions in modernized texts, see Encode Stage Directions in Modernized Texts and Identify Stage Direction Types. For information on encoding stage directions in semi-diplomatic transcriptions, see Encode Stage Directions in Semi-Diplomatic Transcriptions.