Map Of London

We don’t need to mark-up dates in Contributor bios. They function as strings of characters in the bios.

1700 would be a reasonable cut-off for other dates. After 1700, we’re into secondary sources, not primary sources. And whatever happened on a location after 1700 isn’t really “early modern” by the conventions of the discipline of English studies. (Early modern is more capacious for historians.)

The guiding principle needs to be: are we going to do anything with this date? Is it harvestable data, or information that we need to manage the site effectively? If not, then it’s just a string of characters.

As part of fixing the missing text in links in MIDD17, I tested and fixed the system for implementing links from one document to a fragment of another. This is basically how it works:

If you want to link from one document (say MIDD17) to another (say TRIU1), you can link to the whole document using mol:TRIU. However, sometimes you might want to link to a specific part of the TRIU1 document (which is a multi-text document). You can link to a specific <div> in the target document like this:

• First, give the <div> in the target document its own @xml:id. For instance, we might want to link to the "Grocer's Company" section of the TRIU1 document, so we find the div that contains that section, and give it an @xml:id. Our convention is that an @xml:id for a section in a target document should be created with a prefix that consists of the main document id (TRIU1) and an underscore. So we do this: <div xml:id="TRIU1_grocers">.
• In the document where you want to put the link, construct the link as follows: <ref target="mol:TRIU1#TRIU1_grocers">. In other words, mol: + the document id + # + the div id.

Links constructed like this are now working on the site (see MIDD17 links to TRIU1 for examples).

Greetings all,

I am really enjoying my time with MoEML so far. While working through the BIBL1 and PERS1 files I have noticed a few things that we will need to be thinking about in the near (and distant) future. The following is a log of my tasks so far, including notes about what we may need to think about looking forward.

My first task was to delete the dates and names of contributors who have added files to BIBL1 in the past. JJ decided that there was no longer any need for this info. Working in this file, I noticed numerous formatting inconsistencies (arising from different people adding different info at different times with different MLA conventions). I look forward to amending these errors in the coming weeks.

My second task was to ensure that all links in PERS1 were updated. The ODNB had made some changes to their website and so most of our links were broken. Again, I noticed many formatting/style inconsistencies that I am eager to amend in the coming weeks.

My third task was to add the medium (i.e. Print, Web) to each BIBL1 entry. This is a newer MLA convention and has not been used consistently since the website launched. While adding these, I made some changes to the more easily-spottable inconsistencies. This got me pretty excited about a large-scale tidy-up! I am hoping to have this "spring cleaning," as I am calling it, finished by mid-June.

Since I will be spring cleaning for the next few weeks, JJ has assigned me the task of creating an updated style guide for the website. This will ensure that everyone adding information to these files continues to follow a correct and consistent format. I cannot stress enough the importance of consistent formatting for even the most trivial matters like using an en-dash instead of a hyphen between a person's life dates. If we are to continue to assert MoEML as a serious academic publication, we cannot allow formatting errors to persist. I (think I) have attached a draft of the style guide (which is also available in the svn file "documentation," in case the file addition backfired) and I would love to hear your input. It is not yet completely implemented, but please begin referring to it when inputting information. If you have any questions about formatting, please don't hesitate to contact me.

I think that's all for now. Thanks to JJ and MH for all of their guidance so far.

CB

Note to Janelle and to the RAs:

The HCMC computers do not have MS-Word. If we are editing working files, use OpenOffice. If you are on another computer and need to convert a .doc or .docx file to .odt, do NOT use MS-Word to save the file as a .odt file. The comments (where we record so much information for our encoders) will be stripped away in the file conversion.

Instead, save and close the .doc/.docx file. Then, start up OpenOffice and open the .doc/.docx file. Now you can save it as a .odt file without losing the comments.

Remember: File names must not include spaces or punctuation.

JJ requested on Flow that when you click on a person's name, their info be shown in a popup. This is now implemented. Specifically, if you use this type of reference:

<name type="person" ref="mol:HOLM3">Martin Holmes</name>

then the name will generate a popup link, but if you put this:

<ref target="mol:HOLM3">Martin Holmes</ref>

then a link to that person's page will be generated.

First Pass: Preparing a basic text for encoding.

• Copy text from EEBO-TCP print view (this presupposes we have permission to use their transcription -- we're actually using their .xml files for Stow).
• Convert short s to long s (lower case only). (Convert long s before spaces, periods, and commas back to short s.)
• Change the colour of italicized text (for your convenience and the encoder’s convenience). You may want to convert all the text to black and then the italicized text back to a colour once you have finished all your corrections.
• Add in signature numbers as milestones (a way for US to navigate through the book)
• Record what is in the headline (usually the running title but sometimes also a page number) and direction lines (signature number on signed leaves and catchword). EEBO-TCP does not display this info.
• Proofread. Correct errors in EEBO-TCP’s transcription. Correct any long s errors introduced by us. Check punctuation carefully (is it really italicized after names?). Infer gaps in the transcription if you can but don’t make it up if you can’t. If you make an educated editorial guess, put the conjectural characters within editorial square brackets. The microfilms in the library are often clearer.

Second Pass (depends on time and the instructions from JJ).

If you are putting the information into comment bubbles:

• Identify the unique XML:id for all the streets and sites.
• Put notes about queries, unresolved ids, and gaps in database (e.g., people and places who aren't there) into comments. Put a reminder in Flow in Janelle's TTD folder.

If you are printing out the file, highlighting locations, people, bibl items, and other features you want to flag for the encoder:

• Print out file (or send it to Janelle to print out).
• Use pink for bibl tags, yellow for locations, purple for people, green for anything that encoders will need to note (foreign words, book/article titles).
• Look up unique XML:ids and write them on the print-out. If an XML:id is recurrent, you don't need to write it every time unless the context requires you to give it again.

Finally...

• Write a note about the structure of the document. Should we use the simple template (and for most topics and transcriptions we will), the location template, or the complex template? To create sections and subsections and sub-sub-sections, we use the div element. You'll have to think about the structure of the document. Does it consist of sections already? Do we need to divide it into sections? Is there a piece of text that will serve as the header for the section? E.g., The Carriers Cosmographie has letters for section headers, but we also had to make up some editorial section headers. See Basic Document Structure for information. You may need to work out these suggestions by discussing them with the encoders and/or JJ.
• Send file back to JJ.
• When JJ gives the okay, tell the encoders to go ahead. Use Flow to tell them the full name of the file in the SVN working_files folder.