The simplest transcription will do the following:

At the anthology’s discretion, editors may also encode the following:

Semi-diplomatic texts are NOT the place to encode:

Read about the LEMDO Taxonomies: Introduction to LEMDOʼs Taxonomies

LEMDO processes different types of documents in different ways. The processor looks for the <catRef> elements inside the <textClass> element in the <teiHeader> of your document and then applies the appropriate processing for the document types that are captured in the <catRef> elements. If you document does not have <catRef> elements, our processor will not know what type of file it is. If your document has the wrong values in the @target attribute of the <catRef> elements, our processor will apply the wrong processing to your file.

LEMDO has special processing for semi-diplomatic transcriptions (i.e, semi-diplomatic texts), so it is particularly important to get the <catRef> s right for these types of documents.

A semi-diplomatic transcription will have at least four <catRef> elements:

Each <catRef> element has two attributes: @scheme and @target. Use the @scheme attribute to point to the xml:id of the taxonomy; the value must have the tax: prefix. Use the @target attribute to point to the value as defined in that taxonomy; the value must have the cat: prefix.

Optionally, you can add a <catRef> to describe the origin of the document using the LEMDO Document Histories Taxonomy. Semi-diplomatic transcriptions might have come from a converted IML file or from TCP. If this is the case for your file, use the appropriate value: edhSourceIML or edhSourceTCP. If you have transcribed the play yourself into a LEMDO XML template file, you do not need to describe the origin of the document; in this case, you will simply record your witness in the <sourceDesc> element.

Note that <catRef> is an empty element.

Quarto playbook that was converted from an IML-encoded file:

Folio play that was converted from an IML-encoded file:

Quarto playbook that was converted from an EEBO-TCP file:

Signal bibliographical signature numbers (aka editorial signature numbers) in the <pb> element with the @n attribute, as illustrated in the signature number of G2 verso below:

These signature numbers are used by LEMDO processing to generate a table of contents for the semi-diplomatic transcription (except in cases like the Douai Shakespeare Manuscript Project where the labels take precedence over the signature.

All semi-diplomatic texts require tagging to account for columns. Folios generally have two columns. Broadsides may have one, two, or more columns depending on the orientation of the text and the size of the font. Capturing columns helps account for turnovers and turnunders and makes it possible to render the page in columns if we want to.

Use the <cb> element to mark the beginning of each column on a page. Give the cb <element> an @n attribute and numerical value of 1, 2, and so on. Start at 1 on each new page.

Our objective for CSS encoding is to define a set of general rules for representing the text of our semi-diplomatic transcriptions. At the same time, we allow transcribers and encoders to capture the peculiarities of the printed book, both across the entire printed book and at the local level of the line or word. Some knowledge of CSS will help you (the transcriber or encoder), but you do not have to know CSS to have relatively easy control over the way your transcription captures key bibliographical codes and the way it will eventually be rendered.

We follow the principle of minimal tagging for semi-diplomatic transcriptions; LEMDO normally includes links to facsimiles of the books, where people interested in the mise-en-page can see how the page is laid out. We are not interested in exactly recreating the layout of early printed playbooks. We do, however, capture some bibliographical features of the playbook.

LEMDO uses three levels of CSS to capture these bibliographical features:

LEMDO foresees the following scenarios which will determine which level(s) of CSS you will use in your semi-diplomatic transcription:

Printed playbooks are composited and laid out similarly on the printed page. Our default stylesheet captures most key features of this composition and allows for a nice on-screen approximation of an early modern book. In conjunction with our recommended, pre-written file-wide styling, it is sufficient for playbooks that are similar in composition and mise-en-page to most other playbooks, but can also be used in conjunction with more granular styling.

As with all of our encoding, we follow the Endings principles. For styling in semi-diplomatic transcriptions, we must keep the following principles in mind:

Massive redundancy: every page contains all the components it needs, so that it will function without the rest of the site if necessary, even though doing so means duplicating information across the site. (Endings Project, Principles 4.6)

Graceful failure: every page should still function effectively even in the absence of JavaScript or CSS support. (Endings Project, Principles 4.7)

For our semi-diplomatic transcriptions, this means that:

LEMDOʼs default stylesheet will be applied to your semi-diplomatic transcription at rendering time. If it accurately captures the composition and mise-en-page of your playbook or your anthology is not interested in capturing mise-en-page, you do not need to worry about adding any additional CSS other than the recommended, pre-written <tagsDecl> . Our default stylesheet will make your text readable, usable, and approximately bibliographically similar to the original. For our recommened, pre-written <tagsDecl> , see Practice: Add LEMDOʼs Recommended File-Wide Style.

If the HTML output does not look right, check your encoding. For example, if a line is not beginning where it should, check that you have put the <lb> element in the correct place.

If your playbook has deviations from most other playbooks and your anthology is interested in capturing mise-en-page, this default styling will still apply to your semi-diplomatic transcription at rendering time. You can add additional CSS to capture any deviations. See Encode File-Wide Style in Semi-Diplomatic Transcriptions and Encode Inline Style in Semi-Diplomatic Transcriptions.

Stage directions tagged as being in the margin will render in the right margin regardless of whether the @place attribute on <stage> has the value plc-right-margin or plc-left-margin.

LEMDO is not interested in replicating the exact mise-en-page of early modern books. We provide facsimiles for those who are interested in the exact layout of source playbooks. Additionally, the online environment is fundamentally different from early modern printed books—we do not have the same need as early modern compositors to change which side marginal stage directions appear on based on whether a page is recto or verso as we do not have recto and verso pages.

You will still truthfully capture which side of the page marginal stage directions appear on using the value of the @place attribute. For more information on capturing the placement of stage directions in semi-diplomatic transcriptions, see Practice: Encode Placement.

LEMDO has created a default stylesheet that should capture the many key aspects of composition and mise-en-page of most early modern playbooks. There are some components of composition and mise-en-page that must be added in your XML file using the <tagsDecl> element. Additionally, some playbooks deviate from the basic composition represented by our default stylesheet. If your anthology is interested in capturing bibliographical features and your playbook consistently deviates from the features described by our default stylesheet, use the <tagsDecl> element in the <teiHeader> to add CSS across your file.

The <tagsDecl> element allows transcribers and encoders to easily capture key bibliographical features of their playbooks and control how their semi-diplomatic transcription will be rendered. By using it to put file-wide CSS into place, you can exercise this control while maintaining LEMDOʼs principle of including all necessary description of the early modern playbook in the XML file.

Because the placement of forme works is not captured elsewhere in semi-diplomatic transcriptions, you must add the styling for them in the <tagsDecl> element. To do so, copy and paste the following <tagsDecl> element as a child of the <encodingDesc> element in the <teiHeader> of your file: This <tagsDecl> should describe the placement of forme works in most playbooks. Because the position for page numbers tends to alternate between left and right on each page, you will style page numbers using the @place attribute. See Placement Taxonomy for more information.

When writing additional file-wide styling, you will define style that applies universally across your file using the <rendition> element to capture deviations that occur consistently on elements throughout your playbook. This CSS will apply only to elements that you specify using the @selector attribute. For example, you can use CSS to make all the speech prefixes in a playbook look the same. It will apply globally to all the speech prefixes in your text, if you have consistently tagged them with the <speaker> element.

To add a <rendition> element, follow these steps:

For example:

If a consistent bibliographical deviation only occurs in elements with a specific attribute and value on it (for example, if the deviation is in running titles, which are tagged with <fw type="runningTitle">), indicate the attribute and value in the @selector attribute on the <rendition> element. To do this:

For example:

If you would like support in adding style using the <tagsDecl> element, please contact lemdo@uvic.ca to discuss solutions, including the possibility of adding a new <rendition> to our standard set. If you prefer to work out solutions locally using inline CSS, these resources may help:

LEMDO has created a default stylesheet that should capture the basic composition and mise-en-page of most early modern playbooks. Some playbooks deviate from this basic composition. If your playbook sporadically deviates from the features described in our default stylesheet, use the @rendition and @place attributes to capture its features.

Inline styling is the most granular way to capture the features in your playbook. LEMDO uses minimal tagging in semi-diplomatic transciptions, so the inline styling should be used only if 1) your playbook deviates from LEMDOʼs default styling, or 2) the deviations are not consistently on an element that you can select throughout your playbook (i.e., if the deviations are sporadic).

There are two scenarios when applying inline style:

In both of these scenarios, the value on the @rendition attribute will be one of LEMDOʼs preformed renditions. See LEMDOʼs Preformed Renditions.

We use the @place attribute rather than the @rendition attribute to describe the placement of stage directions. Read more about describing the placement of stage directions in semi-diplomatic transcriptions in Practice: Encode Placement.

Stage directions have unique default styling. For more on how stage directions in semi-diplomatic transcriptions render, see Rendering.

LEMDO has created a number of rendition values. LEMDO has defined the CSS that applies to each of these values so that you can use them without needing to learn CSS. See Renditions Taxonomy.

If LEMDO does not have a preformed rendition that describes the deviations in your text, please contact lemdo@uvic.ca to discuss solutions, including the possibility of adding a new <rendition> to our standard set.

If you are experienced with CSS, you may choose to either:

Example of the @rendition attribute on a container element:

Example of the @rendition attribute on the <hi> element:

Example of the @place attribute on the <stage> element:

LEMDO has created a default stylesheet that should capture the basic composition and mise-en-page of most early modern playbooks. Some playbooks deviate from this basic mise-en-page in consistent ways, in which case you can use anthology-level CSS or create renditions in the <teiHeader> of each individual file. Inline deviations from the anthology-level CSS or file-level renditions can be described using inline CSS. If your playbook sporadically deviates from the features described in our default stylesheet in a way that cannot be captured by our pre-fabricated renditions or by regular CSS, use advanced CSS options such as CSS Flex.

Only use the advanced CSS options outlined in this documentation if: If you think that your file requires advanced CSS but you do not have the experience necessary to use it, please contact lemdotech@uvic.ca.

CSS Flex allows us to truthfully encode and accurately render features that cannot be captured by standard CSS.

CSS typically uses block formatting, which renders an element on the line below the previous element, or inline formatting, which renders an element on the same line as the previous element. CSS Flex allows for elements that do not fit either the block or inline format and are not tables. For example, although we could potentially encode long braces using a table with styling, that method would not be a truthful depiction of the page layout in early modern playbooks.

We primarily use CSS Flex for inline long braces. There may be other features that are best captured by CSS Flex. If you come across a different use case for CSS Flex in your edition, let the LEMDO team know by emailing lemdotech@uvic.ca.

The first step to encode long braces is to break down the relevant compositorial units on the page that you are encoding. You have to think like an early modern compositor. What are the basic blocks on the page that you need to be able to style in some way? Each thing that needs to be styled (stretched, centered, italicized, grouped) needs to be contained in its own element.

If you are encoding a list wrapped in a pair of long braces, your components might be: list heading ( <label> ), brace ( <metamark> ), list ( <list> ), item(s) ( <item> ), and brace ( <metamark> ). Each of these components will be represented as an element in TEI. If there is no obvious element to act as a container for the component, you may use the <hi> or <seg> element.

Next, group all of the components that you have identified by wrapping them in a <seg> element. Put a @style attribute on the <seg> . There are three components included in the value of @style. Format these as you do for typical CSS (i.e., put a semicolon after each component and separate components with a space). The first component is display. Because we are using CSS Flex, we want to have the value display: flex;. The second component is flex-direction. You have the option to use either flex-direction: row; or flex-direction: column;. If you choose row as your flex direction, content will be centered in vertical space; if you choose column as your flex direction, content will be centred in horizontal space. The third component is align-items. We typically want items to be centred when we use CSS Flex, so use align-items: center; (note American spelling in CSS values).

You should now have something that looks either like this: or like this:

To stretch your braces, add a @style attribute to the <metamark> element. There are three components to the value for @style. The first component is transform, which defines the axis that you want to stretch. For long braces, you will stretch the vertical or Y axis. To do this, encode as transform: scaleY followed by a number equivalent to the number of sizes larger that you want the element to stretch wrapped in parentheses (for example, to stretch a brace to three times its normal height, the transform component would be encoded as: transform: scaleY (3);). The second component is transform-origin, which defines where the stretch starts. We typically want this to be transform-origin: left-top;. The third component is margin, which contains four numerical values seperated with spaces. The first value corresponds to the top, the second to the right, the third to the bottom, and the fourth to the left. You will need to tweak these values until your braces look correct. Note that after values other than zero, you will need to specify em as the unit after the number. If the value is 0, you do not need to specify the unit.

This is an example of a brace encoded to be three times its original height and transformed to have a margin of -2.4 ems from the top and 0.2 ems from the left:

You will likely need to tweak the margin values. See for options for checking and tweaking your CSS Flex values.

There are three ways to check and tweak your CSS. The first is to commit your file to the LEMDO repository, wait for it to build on the website, and use inspector view to change the CSS. The second is to use a computer that has Ant installed and create a local build. The third is to use a coding site such as CodePen. The first option works well as long as there is at least one successful build of your page on the Jenkins site. The second option works even if there is not yet a successful build. The third option requires at least one successful build of your page on the Jenkins site, but shows immediate results from tweaks to CSS and has a clearly laid out code editor.

To tweak your CSS in inspector, follow these steps:

To check and tweak your CSS using a local build, follow these steps: Note that not all JavaScript will work correctly when you download the static site, but the layout and styling will appear. You will not need the JavaScript for tweaking your CSS.

To tweak your code with CodePen, go to the Pen feature of the CodePen website, then follow these steps:

This example shows how to apply CSS Flex to a stretched brace used to group characters that share a role description in a manuscript cast list. The parent element containing all the boxes is the first <castGroup> element. The first box is the child <castGroup> element. The second box is the <metamark> element containing the brace (French: accolade), which has styling to stretch the brace. The third box is the <roleDesc> :

The following resources have helpful information and tools for learning and practicing CSS Flex:

The <docTitle> element contains the full title of the playbook. Titles of playbooks are often long, with extended descriptions of the action of the play and possibly information about performance. Wrap the entire title in the <docTitle> element.

The <docTitle> element must include one or more <titlePart> elements. LEMDO has adopted the TEI recommended values for the @type attribute on <titlePart> : main (main title), sub (subtitle), alt (alternate title), and desc (descriptive title). However, there is no one pattern we can impose on early modern title pages, and, at the moment, we do not have plans to process or query the parts of title pages. If we were a project interested primarily in title pages, we would likely identify multiple title page patterns and prescribe encoding models for each one. Our advice to editors: use the values that make the most sense for the parts of the title of your playbook, in consultation with your anthology lead. Our advice to Encoders and Remediators on the LEMDO team: do your best with the knowledge that you have of the play (remembering that title pages double as advertisements for the playbook), ask the editor of the edition (if the editor is still active in the field and available for questions), or ask the Project Director.

Not all title pages will have all four parts. Some titles will have two or more parts that you might want to tag with the same value. For example, you might decide to treat all the parts of the title except the main title as desc. In other words, there is no requirement to use all four <titlePart> values, nor is there a prohibition against using a value more than once.

Wrap the components of the title in <titlePart> elements. Add a @type attribute to each <titlePart> with one of the following values: main, sub, alt, desc.

Punctuation is part of the title. Include the punctuation within the text node of <titlePart> .

¶ The Main Title

¶ Subordinate Titles

¶ Descriptive Titles

¶ Alternate Titles

The byline on a playbook title page is usually short. Wrap any text that indicates authorship (whether accurate or spurious) in the <byline> element. Include Terminal punctuation inside the <byline> element. Do not tag the names; the LEMDO team will tag names in bylines before we publish your transcription.

Some playbooks include an epigraph above or below the byline. The <epigraph> element cannot contain text alone. Wrap the text in an <ab> element and use the milestone <lb> element to capture compositorial line beginnings in the epigraph. Note that although epigraphs are often in a foreign language, they should not include the <foreign> element. LEMDO does not tag foreign languages in semi-diplomatic transcriptions. Instead, simply add a @rendition attribute with a value of rnd:italic if the text is in italic font.

Figures include fleurons, ruled lines, and printerʼs ornaments. Because LEMDO users will always have access to the digital surrogates of the playbooks, we do not try to reproduce these figures. Instead, use the <figure> element to indicate where a figure occurs, and the <figDesc> element to describe it using the following terms (a LEMDO controlled vocabulary):

If these terms do not adequately capture what appears on your title page, contact lemdo@uvic.ca to discuss the possibility of adding a new term.

Do not put an <lb> element in front of your figure. We do not count figures as lines in LEMDO. At rendering time, LEMDO will add padding before and after figures.

Use the <docImprint> element to capture all the information on the title page about where, by whom, and when the book was created and made available. Transcribe the entire imprint line within the <docImprint> element and use child <persName> , <pubPlace> , and <docDate> elements to capture specific entitites contained therein and make them processable. The LEMDO team will add the printer to the historical prosopography (PROS1) and make a link later so that the final encoding will look as follows:

¶ Place of Publication

¶ Date of Publication

¶ Practice: Encode the Names of Printers, Publishers, and Booksellers

Below is a full example of an encoded title page:

By front matter, LEMDO means one of the following: Our taxonomy of front matter is a subset of the taxonomy of paratexts developed by Sonia Massai and Heidi Craig for the Early Modern Dramatic Paratexts project.

While some digital documentary editions exclude some or all of the front matter, LEMDO considers the entire playbook to be significant.

Title pages require more complex encoding than other types of front matter. See Encode Title Page of Semi-Diplomatic Transcriptions.

In the case of plays in collections (e.g., the 1623 folio of Shakespeare’s plays), the anthology containing those plays will decide on how to handle the front matter of the collection.

A few types of paratext can appear in either the front matter or the back matter.

All of the front matter belongs inside the <front> element (a child of <text> ). Wrap each discrete section of front matter in its own <div> element. Add the @type attribute and the appropriate value. Wrap the title of the section in the <head> element. If there are subheadings in the section, wrap them in <label> and style with CSS as needed.

Labels such as Finis are not considered back matter in semi-diplomatic transcriptions. See Encode Literary Divisions in Semi-Diplomatic Transcriptions.

By back matter, LEMDO means one of the following: Our taxonomy of back matter is a subset of the taxonomy of paratexts developed by Sonia Massai and Heidi Craig for the Early Modern Dramatic Paratexts project.

While some digital documentary editions exclude some or all of the back matter, LEMDO considers the entire playbook to be significant.

In the case of plays in collections (e.g., the 1623 folio of Shakespeare’s plays), the anthology containing those plays will decide on how to handle the back matter of the collection.

All of the back matter belongs inside the <back> element (a child of <text> ). Wrap each discrete section of back matter in its own <div> element. Add the @type attribute and the appropriate value. Wrap the title of the section in the <head> element. If there are subheadings in the section, wrap them in <label> and style with CSS as needed.

Treatment of back matter is similar to LEMDO’s treatment of front matter. See Encode Front Matter in Semi-Diplomatic Transcriptions.

We do not impose act and scene divisions on semi-diplomatic texts. Most witnesses (folios, quartos, and octavos) have few or inconsistent act and scene numbers. The semi-diplomatic text is meant to be a conservative documentary edition and is therefore not the place to make critical decisions about when scenes break.

Likewise, we do not give structural significance to the common ending signifier Finis, in part because a number of early publications (including the Q1 Pericles) have multiple instances of Finis.

Transcribe the act and scene headings that do appear in the witness, as well as phrases like Finis. Treat them not as literary divisions but as labels. Tag them with the <label> element and the @type attribute with the value heading.

Depending on your anthologyʼs practice, you may use the self-closing <milestone> element with the @unit attribute and the value act or scene to indicate the beginning of the literary division. You may also add an @n attribute and a value consisting of a number. In the case of the first scene of the first act, place the <milestone> element after the title of the play and before the act/scene heading. Use the <label> element for the heading.

These examples contain some inline CSS styling. See section on Rendering below.

LEMDO does not have default styling for <label> in part because there are so many places in which label-like things can appear but also because printers tend to exercise some creativity with labels. You almost always need to add CSS styling for labels, either in your TEI header (if the things you are tagging with <label> are consistently styled across your document) or inline.

In cases where a witness does have consistent and inclusive act and/or scene boundaries, you will follow the same practice as above. Do not treat these cases any differently. Use <milestone> elements (not divisions) and <label> for the header.

¶ Examples

We do not add proleptic act and scene divisions to our semi-diplomatic texts. Acts and scenes are an editorial imposition. Even now, they are subject to editorial emendation and differ from one edition to another. For example, the scenes often numbered 2.1 and 2.2 in editions of Romeo and Juliet are treated as one scene in the Oxford Shakespeare (and the Norton 1 and Norton 2 based thereon).

However, your anthology may choose to add anteleptic milestones to the semi-diplomatic transcription once a modern text based on this copytext is stable. Doing so may facilitate future linking between the semi-diplomatic transcription and the modern text based thereon.

In this case, the optional practice is to add <milestone> elements at the points where the act and scene divisions open in your modern text (i.e., the opening <div> tag in your modern text). Add a @resp attribute to the <milestone> element. The value of @resp will be the pers: prefix plus your own xml:id. Add a @corresp attribute that points to the modern document and the xml:id of the <div> to which you want to point.

The <castList> element is used only for character lists that appear in early modern playbooks (manuscript or print). Do not add a character list to a semi-diplomatic transcription unless there is already a character list in the document. LEMDO does not use the <castList> element for editorial character lists in modern texts; instead, the editor will create a <listPerson> in the <teiHeader> of the modern text.

The print or manuscript list of characters is part of the text that you are transcribing in a semi-diplomatic transcription. It contains important information for book and performance historians. The list of characters either precedes (as in the case of Rhodon and Iris) or follows (as in the case of Timon of Athens in F1) the body of the play.

Placement: The <castList> belongs in either the <front> element if it precedes the spoken text of the play, or in the <back> element if it follows the spoken text of the play. Where there are other preliminary or postliminal materials, encode the <castList> where it appears in the sequence of paratexts in the copy you are transcribing.

Wrap the entire character list in the <castList> element. If there is a header (e.g., “Dramatis Personae” or “The Names of the Actors”, transcribe it accurately and wrap it in a <head> element as the first child of <castList> . The basic model for encoding each character is as follows:

If characters are grouped together (e.g., with a brace) so as to share a <roleDesc> , use the <castGroup> element to group two or more <castItem> elements. The <roleDesc> goes outside the <castItem> elements. The basic model for encoding a character group is as follows:

If you want to capture the mise-en-page and the length of the brace using CSS Flex, you will need to wrap an additional <castGroup> element around your <castItem> elements and add the <metamark> element for the brace. The basic model for encoding a character group with the brace is as follows: This encoding pattern gives you three cognate containers to style using CSS Flex: the child <castGroup> element, its sibling <metamark> element, and its sibling <roleDesc> element. See Encode Advanced Style in Semi-Diplomatic Transcriptions.

Optional: You may assign an xml:id to the <castItem> element if you want to associate <sp> elements with characters (using the @who attribute and the xml:id you assigned to the <castItem> .

LEMDO wants to faithfully capture text in source playbooks. For that reason, we wrap the <sp> element around single speeches as they are composed in the source books. We do not supply who delivers a speech—it is the editorʼs job to assign this in the modern text only.

To encode speeches, follow these steps:

Note that we do not use the @who attribute in semi-diplomatic transcriptions.

For example:

To number each speech, add an @xml:id attribute on the <sp> element. The value of the @xml:id must be prefixed by the name of the file followed by an underscore, sp, and sequential numbers.

Our Schematron will prompt you if you skip a number in the sequence. Our Schema will also catch any repeated values when you validate your file.

For example:

You may come across a speech that spans multiple lines floating within the compositorial block beside other speeches or stage directions in your source playbook. If you come across this scenario, you will put the @rendition attribute with the value rnd:rightAdjacent on the <sp> element. Do not add an <lb> element before the opening <sp> tag. Do add <lb> elements at the start of each subsequent line within the speech, but do not add the @type or @n attributes on these <lb> elements.

For example:

LEMDO treats stage directions as a quasi-paratextual feature of the text and as a function of the mise-en-page. We encode the type of stage direction using the @type attribute in anticipation of a day when we will be able to generate a database of stage directions. We capture the placement of the stage direction on the page using the @place attribute so that we can render the stage direction in a similar place to where it appears in the early modern playbook.

Stage directions provide information about what happens on stage, where the scene takes place, time of day, character relationships, status, and much more. LEMDO has created a list of allowed values on the @type attribute to specify the function of the stage direction.

To encode @type values, first read through the entire stage direction and parse out the functions of any component parts. Next, add a @type attribute and select the relevant value(s) from the dropdown menu in Oxygen. Many stage directions are simple and need only one value on the @type attribute. Some stage directions are more complicated and need multiple @type values. If a stage direction in your file is more complex and needs multiple values, type the first value followed by a space and then a second value, and so on (i.e., @type="value1 value2 value3").

LEMDO has built upon the TEI definitions for @type values and expanded them based on our project-specific needs for working with early modern dramatic works. See the table below for detailed information on each @type value in the LEMDO project:

Value	LEMDO Definition
setting	Describes a setting (e.g., in her bed, a council chamber). Do not use this value to describe a location on the stage.
location	Describes a location on stage (e.g., above or at one door). Do not use this value to describe a setting. Do not assign a location that is not required by early modern stage direction or implied by a dialogic stage direction.
entrance	Marks the entrance of one or more characters.
exit	Marks the exit of one or more characters.
business	Describes stage business and character actions like kneeling. Use for stage directions marked in the playbook as dumbshows.
delivery	Describes how or to whom a character speaks, normally onstage but allowed for offstage stage directions (e.g., they cry within).
sound	Describes a sound such as flourish, music, thunder, a shot, drums, whistle, whether the sound is made onstage or offstage.
remain	Indicates that one more more characters remains on stage when others exit. Old IML value was usually other. This is a LEMDO project-specific value.

Our dropdown menu in Oxygen also allows optional and other as values on the @type attribute. However, these are legacy values that we are depricating. Do not use these values. Instead, use the more specific values as outlined in the table above.

To describe the placement of a stage direction on the page, add a @place attribute with a value from LEMDOʼs placement taxonomy. Consult with your anthology lead to determine if your anthology is using generic (e.g., plc-top) or specific (e.g., plc-left-top) placement values.

Read more about the placement taxonomy and see pages annotated with examples of @place values in Placement Taxonomy.

If a stage direction is inside a speech and on the same compositorial line as a line of the speech, tag it with <stage> as usual. If the stage direction has parentheses around it, type the parentheses and include them inside the <stage> element.¹

Do not add an <lb> unless the stage direction is on its own line within the speech.

Use the @place attribute to indicate the location of the stage direction on the line as you typically would.

If your source playbook has marginal stage directions, follow these steps:

To learn about how marginal stage directions will render, see Special Case: Renditions for Marginal Stage Directions.

Our generic stylesheet for semi-diplomatic transcriptions automatically renders all stage directions in italic font. Some stage directions in your playbook may be entirely or partially composed in roman font. If your anthology wants you to capture font features such as italic fonts, follow the steps outlined below.

If your stage direction is entirely in roman font:

If your stage direction is partly in roman font:

To learn more about our stylesheet for semi-diplomatic transcriptions, see Default Style in Semi-Diplomatic Transcriptions.

Example of the final exit in a scene, placed on its own line and aligned right. The first letter of Exeunt is not italicized.

Example of a stage direction centered on a new compositorial line.

Example of a stage direction aligned right on the same compositorial line as the last line of the speech.

Example of a stage direction with multiple @type values: Even though this is a relatively short stage direction, there are three different @type values.

Example of a marginal stage direction:

We do not use the <lg> (line group) element, <l> (line) element, or <p> (paragraph) element in our semi-diplomatic transcriptions. Compositors sometimes set verse as prose to save space, or prose as verse to fill up more page space. We do not make any claims in the semi-diplomatic transcription about whether a semi-diplomatic text is verse or prose.

Use the milestone <lb> element to demarcate the beginning of a compositorial line. When the compositorial line begins with the speech prefix, put the <lb> element before the <sp> element. See Practice: Encode Speeches.

If a compositor breaks a word across a line, type the hyphen and wrap it in the <pc> element. Add the @force and the value weak.

The reason we ask you to encode these is that our static search engine is programmed to ignore tagged hyphens so that it can find words that are broken across lines.²

All LEMDO semi-diplomatic transcriptions are given witness line numbers (WLNs). Add the @type attribute to the <lb> element that begins the compositorial line with the value of wln. Add an @n attribute with a numerical. Numerical values must be consecutive whole numbers. Do not use decimals.

On the question of which compositorial lines are numbered, LEMDO follows Charles Hinman’s practice for TLNs in The Norton Facsimile (Hinman xxiii-xxiv). The first countable WLN is the first compositorial line after the title. Usually, the first compositorial line after the title will be an introduction such as ACTUS I or similar. It is important to remember that act and scene numbers in these texts are inconsistent, so there may be times when the first compositorial line after the title is not an act or scene number but rather the opening stage direction, or the first speaker element. In these cases, you must still tag <lb type="wln" n="1"/> on the first compositorial line after the title. If the placement of the first WLN is incorrect in your file, then all subsequent WLNs will need to be revised. Reassigning WLNs is a process, so if you are unsure where the first compositorial line is on your Old-Spelling text, check in with a project lead to get confirmation before proceeding.

In summation: lb type="wln" n="1"/ must appear on the first compositorial line after the title of the play, regardless of what is on that compositorial line.

Neither forme works nor vertical spaces are encoded using <lb> . Therefore, neither receives WLNs.

If there are marginal stage directions in your source playbook, add an <lb> element at the beginning of each line regardless if they appear inline with lines in the compositorial block. Do not put the @type or @n attributes on these <lb> elements. As marginal stage directions are not part of the compositorial block, they are not included in the sequential numbering of compositorial lines.

For detailed practice on encoding marginal stage directions, see Special Case: Encode Marginal Stage Directions.

If you come across an interjection, simultaneous speech, or floating stage direction that spans multiple lines and is to the right of a speech or other literary division and is within the compositorial block (i.e., not in the margin), do not give the <lb> s in that floating block @type or @n attributes. They share a compositorial line with other text, which will have numbered WLNs. To learn more about these cases, see plc-right-adjacent in Placement Values and Placement Value Examples.

At rendering time, LEMDO breaks lines according to the placement of the <lb> elements in your encoding. On very narrow screens, long compositorial lines may wrap dynamically but there will always be a break wherever you have captured an <lb> .

When you are transcribing your semi-diplomatic text, type the hungword in the semantic line to which it belongs. Tag turnovers with the <seg> element thus:

Tag turnunders with the <seg> element thus:

When we render these lines, we will push the hungwords above or below the line as needed.

Some hungwords appear on an otherwise empty line.

If you come across this scenario, encode the hungword as usual. Encode the white space as vertical white space. See Practice: Encode Vertical White Space.

For example:

We use the <space> element in our semi-diplomatic transcriptions judiciously to capture vertical space between printed units and to capture horizontal space within lines if it is semantically significant.

If there is white space with no type or ornaments between lines, use the <space> element. Add the @dim attribute with the value vertical, the @unit attribute with the value line, and the @quantity attribute with the numerical value equivalent to the number of lines of white space present in your source.

If there is horizontal white space, use the <space> element. Add the @dim attribute with the value horizontal, the @unit attribute with the value char, and the @quantity attribute with the numerical value equivalent to the number of ems that are white space present in your source.

Note that we silently modernize spacing between words because it is too difficult to exactly replicate the various spacing between words. We only encode horizontal white space if it is of semantic significance. People interested in the exact spacing between words can view our facsimiles.

Example of vertical space:

Example of horizontal space:

Signature marks were meant to help the book binder assemble printed sheets in the right order. Early modern printed books were constructed out of folded sheets of paper. Normally, a book printed in folio is made from gatherings of three folded sheets (a folio in sixes). Each sheet is folded in half and the three sheets are then nested or quired.³ Normally, a book printed in quarto is made from one sheet of paper folded twice (a quarto in fours) or from two quired twice-folded sheets (a quarto in eights).

To help the bookbinder fold and quire these sheets together in the right order to make a book, each sheet is marked with a signature mark. The signature mark is an alphanumeric string appearing at the bottom of what will become a page when the sheet is folded and bound. A gathering marked with A on the first sheet, A2 on the second, and so on is called the A gathering. A gathering marked with B on the first leaf, B2 on the second leaf, and so on is called the B gathering.

The printer uses 23 letters in these alphanumeric strings: the standard English alphabet but with only one of I or J (interchangeable characters in early modern English) and only one of U or V (also interchangeable in early modern English); W is omitted because of the potential confusion with VV. For long books that require more than 23 gatherings, the printer goes through the alphabet again, like this: Aa, Bb, and so on. A third set of gatherings would be numbered Aaa, Bbb, and so on.

When the sheets are folded, the signature marks appear at the bottom of some pages. They are never on the verso (second side) of a leaf, which means they appear only on the right side of an opening (the two-page spread that you are looking at when you read a book). They usually do not appear on the conjugate leaves (the second leaf of a folded sheet in the case of a folio, or the two leaves that remain connected in the final binding regardless of folding and cutting).

The following table gives the most common patterns for signature marks in folio-size books. Because the A gathering is often the title gathering (with an unsigned title page and fewer quired sheets), this table describes a typical B gathering:

The following table gives the most common patterns for signature marks in quarto-size books, again taking the B gathering as an example:

See Deciphering Signature Marks by Sarah Werner on the Folger Shakespeare Library’s The Collation blog.

To see an array of signature marks on early printed pages, see the search results at Main Printed Feature: Signature Mark on Sarah Werner’s Early Printed Books: Resources for Learning and Teaching.

For a video introduction to the anatomy of books (including signature marks), see Anne Pealeʼs Key Terms in Book History on the University of Edinburghʼs Centre for the History of the Book Instructional Videos page.

For a video guide to book format (quartos, folios, etc.), see Elizabeth Quarmby-Lawrenceʼs Bibliographical Formats on the University of Edinburghʼs Centre for the History of the Book Media Content page.

For a fun paper-folding exercise, download the practice early modern folding sheet from the Folger Shakespeare Libraryʼs DIY Quarto page.

Running titles are encoded as forme works using the <fw> element and the @type attribute with the value runningTitle. Supply running titles with the <supplied> element only when the number is cut off, fuzzy, or over-inked. The following is a typical running title:

The default rendering of anything tagged with <fw> and the runningTitle value on the @type attribute is italicization. You can override our generic styling in a number of ways; for information on adding styling to your text, see Introduction to Style in Semi-Diplomatic Transcriptions. The default placement for anything tagged as a running title is centre top. If your anthology wants you to capture the composition of the page and a running title is not in the centre top of a page, add a @place attribute with a value from LEMDOʼs placement taxonomy. Read more about placement and see examples of placement values in Placement Taxonomy.

If your anthology wants you to capture font features like italics, note that running titles are sometimes only partially set in italic type or partially set in roman type. If a running title is fully roman, add a @rendition attribute to the <fw> element with the value rnd:normal. If the running title is partially in roman type, wrap a <hi> element around the part that is in roman and add the @rendition attribute to the <hi> element. Read more about encoding inline style in semi-diplomatic transcriptions in Encode Inline Style in Semi-Diplomatic Transcriptions.

Explicit signature marks are encoded as forme works using the <fw> element and the @type attribute with the value sig. Infer and supply signature marks with the <supplied> element only when the number is cut off, fuzzy, or over-inked. The following is a typical signature mark:

You may close up spaces in your encoding between the letter and the number if the witness has spaces, unless your anthology lead tells you otherwise. For example, if the leaf is numbered A 2 in the forme work, encode it as A2.

If your anthology wants you to capture the composition of the page in full and the signature mark in your witness has spaces, close up the space between the letter and the number in your transcription (for machine-reading purposes) but add a @rendition attribute with the value rnd:letterspace to indicate how the signature mark has been composited:

LEMDO has added default styling that places signature marks in the centre bottom of each page. If your anthology wants you to capture the composition of the page and a signature mark is not in the centre bottom of a page, add a @place attribute with a value from LEMDOʼs placement taxonomy. Read more about placement and see examples of placement values in Placement Taxonomy.

If the page is not signed, there is nothing to capture in the <fw> element. Inferred or bibliographic signature numbers are added to the <pb> element. See Introduction to Signature Marks.

Explicit catchwords are encoded as forme works using the <fw> element and the @type attribute with the value catch. Infer and supply catchwords with the <supplied> element only when the word is cut off, fuzzy, or over-inked. The following is a typical catchword:

Special case: Catchwords may be in italic type if the word is italic on the next page (e.g., if it is part of a character name or a stage direction). To encode italicized catchwords, add the @rendition attribute with the value rnd:italic on the <fw> element. Read more about encoding style in semi-diplomatic transcriptions in Introduction to Style in Semi-Diplomatic Transcriptions.

LEMDO has added default styling that places catchwords in the right bottom of each page. If your anthology lead wants you to capture the composition of the page and a catchword is not in the right bottom of a page, add a @place attribute with a value from LEMDOʼs placement taxonomy. Read more about placement and see examples of placement values in Placement Taxonomy.

Page numbers refer to the printed numbers in the original playbook, not to the signature marks or inferred signature numbers. Explicit page numbers are encoded as forme works using the <fw> element and the @type attribute with the value pageNum. Supply page numbers with the <supplied> element only when the number is cut off, fuzzy, or overinked. Signal where the page number appears using the following @place values: plc-right-top or plc-left-top:

LEMDO does not make editorial claims about type of text (e.g., whether text is verse or prose) in semi-diplomatic transcriptions. Because of this principle, we do not have special encoding practice for letters and songs in semi-diplomatic transcriptions. This documentation will guide you through using our standard speech encoding practice in the context of letters and songs in semi-diplomatic transcriptions.

Generally, you will encode letters and songs as you encode speeches in semi-diplomatic transcriptions. However, there are some special compositorial features that are common to letters and songs. To encode these features, follow these steps:

For example:

Occasionally, you will encounter obvious compositorial errors in the witness you are transcribing. These errors might be typographical (e.g., if a piece of type has been misdistributed into the case) or derive from a misreading of the manuscript. If your anthology wants you to correct these errors, LEMDO has an encoding protocol that allows you to capture the error and your correction thereof.

Do not use this method to capture press variants. If your anthology wants you to do a horizontal collation of press variants, you will need to create a collation file for your semi-diplomatic transcription. See Collation Types.

Transcribe the error and wrap it in the <sic> element. Provide the correction and wrap it in the <corr> element. Put the parent <choice> element around the sibling <sic> and <corr> elements.

You may encounter instances in your playbook in which some letters appear rotated. To encode turned type, see Encode Rotated Letters in Semi-Diplomatic Texts.

LEMDO recommends that semi-diplomatic transcriptions be prepared from a single material witness and link to facsimiles of pages in the witness. Our suggested criteria for selecting that witness are listed in Select Images for Semi-Diplomatic Transcriptions. LEMDO does not require a horizontal collation of variants but it does provide encoding mechanisms for capturing press variants. You and your anthology lead(s) will make the call about whether or not your edition will include press variants and which method to deploy.

If you are confident that you know which sheets in your witness represent the uncorrected state and which represent the corrected state, you may use the <choice> element with child <sic> and <corr> elements to capture the corrected readings. Note that this method presupposes that the witness you have transcribed contains an uncorrected sheet with readings that you wish to emend after consulting a witness with a corrected sheet. You would not use this method to do a complete collation of press variants. If your witness contains the corrected sheets and you want to capture the reading in uncorrected sheets in another witness, use a collation file instead, as described in the second practice.

In the XML file containing your semi-diplomatic transcription, wrap the uncorrected character, word, or string in a <sic> element. Supply the corrected character, word, or string in the <corr> element. Wrap the sibling elements in a parent <choice> element. Add a @source attribute to the <corr> element. The value of the @source attribute will be the xml:id of the source from which you take the corrected reading.

¶ Examples

The objective of our semi-diplomatic transcriptions is to provide faithful representations of the text in source playbooks. For that reason, we supply materials sparingly in semi-diplomatic transcriptions. Only supply emendations if there is material that is:

Do not use the <supplied> element to correct compositional errors or to expand text that is abbreviated in the material that you are transcribing. See Encode Corrections in Semi-Diplomatic Transcriptions and Encode Abbreviations in Semi-Diplomatic Transcriptions.

To encode emendations, wrap your supplied text in the <supplied> element. Add the @cert (certainty) attribute with a value from the drop-down menu (low, medium, high, or unknown) to indicate the certainty of your emendation. Add the @reason attribute with a value from the drop-down menu to indicate why you are supplying material. Add the @resp attribute to give yourself credit for the emendation. Add the prefix pers: followed by your xml:id as the value of @resp.