Upgrade StorageNeed to italicize part a title
Frequently in the humanities, the title of a book appears within the title of a journal article. The book title needs to be italicized in the citation or bibliography. E.g., if an article is titled, "Moral Obligation in Milton's Paradise Lost," the title "Paradise Lost" needs to appear in italics. If this is currently possible, or someone has a workaround to implement it, please comment. If not, then this is a feature request.
In addition, subscript, superscript, and greek characters occur frequently within the title of references.
Using Unicode wherever possible is good (IMHO), for the other stuff (such as italics) I'd prefer a simple markup language (which could be also used for notes). Markdown comes to mind here, unfortunately Markdown does not support sub/super-script, which would be nice to have for sub/super-scripted letters. Another approach would be to use semantic markup, which bdarcus was probably referring to. I'd like that approach as well.
But that gets to questions of input/output. How does a user encode that information, and how is it preserved on export and so forth?
On the latter, I see two options: some wiki-markup-like subset where the title or species name is still just string (a simple RDF property), or just encode it in XHTML as an XML literal in the property, where you just have spans with classes like "title" and "species-name" and so forth. My hunch is the second is better and more flexible.
The issue comes up, as we've noted, with titles within titles and species/genus names, but also foreign words and the names of ships.
Chicago makes it clear that how these are treated typographically depends on the context. For example, sometime a title within a title is italicized, but sometimes it should be left roman and placed inside quotation marks.
So if a string had a tag such as "title", "shipname", "foreign" or "speciesgenus" it could get formatted appropriately when used in footnotes or bibliographies (and their various versions) and differently when being output to a database.
I guess that means inputting can create an insurmountable problem. If a string in a title has quotation marks, we can't know whether the enclosed string is a title or something that should remain as a string with quotation marks (if in fact there can be such cases).
In the short-term, I am putting and in my titles and have added the following lines to the ZoteroInsertFormattedText function in zotero.dot (right after tagIndex = -1).
While InStr(text, "")
backslashIndex = InStr(text, "")
text = Left$(text, backslashIndex - 1) & "\i " & Mid$(text, backslashIndex + 3)
Wend
While InStr(text, "")
backslashIndex = InStr(text, "")
text = Left$(text, backslashIndex - 1) & "\i0 " & Mid$(text, backslashIndex + 4)
Wend
It's a hack, but it seems to get me out of a jam. I'm trying to use Word and Zotero to write an actual publication-ready article for a history-of-philosophy journal and seeing if I can push through any obstacles. If I couldn't find a workaround on this, I was going to be stuck back with EndNote.
Thanks for Zotero and for listening to feedback from early users! Congratulations.
That should have said:
In the short-term, I am putting <I> and </I> in my titles and have added the following lines to the ZoteroInsertFormattedText function in zotero.dot (right after tagIndex = -1).
While InStr(text, "<I>")
backslashIndex = InStr(text, "<I>")
text = Left$(text, backslashIndex - 1) & "\i " & Mid$(text, backslashIndex + 3)
Wend
While InStr(text, "</I>")
backslashIndex = InStr(text, "</I>")
text = Left$(text, backslashIndex - 1) & "\i0 " & Mid$(text, backslashIndex + 4)
Wend
<dc:title>A <i>title</i></dc:title>
... which won't be valid. Come to think of it, it would probably break validity for any XML export.
If that might be an issue for you, it would be safer to use a non-XML solution like the wiki-markup _title_?
Good luck.
For semantic markup, I think using span/class makes the most sense. Ideally, I'd like to solve this problem in a way that kept users from needing to enter the markup by hand (and instead, say, highlighting a block of text and using context menu options prepopulated with various semantic descriptors), but that would be considerably more involved to implement and, frankly, I've never seen WYSIWYG markup editing that actually really worked after the first edit. (At the very least, if we did do that, we would still need a manual mode.) It might be nice to have that as a goal, especially for notes, but to start, at least, this will probably just be by hand.
Some questions:
1) Is there a fixed set of semantic descriptors ('title', 'species-name', etc.), and/or is there a generic fallback style for unknown classes?
2) Do we still support <i> and other presentational HTML tags in metadata fields?
2.5) At least for notes, we probably do need to support WYSIWYG HTML. How is semantic markup handled in notes, and what special behaviors (linking to other items, etc.), if any, does it have? Bruce began to talk about this a little on another thread. This is a non-trivial problem and ultimately is probably related to the sorts of URI issues that Bruce brought up on the dev list.
3) To what extent are the CSL style rules followed within Zotero itself? I don't think the items list even supports rich text (at least, granularly--I think you can just style a whole cell), so that's probably out. The metadata table could be made to work, but it might make sense to just default to italics for any <span>, since we probably don't want to run every field through the CSL engine, and, of course, the formatting would be style-specific.
Semantic metadata fields and HTML in notes should be able to happen pretty soon once we resolve these issues. Truly semantic/relational markup in notes is a more difficult problem and is probably a bit further down the line.
To answer your questions:
1) The most common issues I'm aware of in titles are math markup (though David Carlisle once told me he thought MathML was not needed for titles, and that unicode could cover that; not so for abstracts), inline titles, phrases/quotes, latin species names.
So I think it is possible to identify a small handful of semantic terms, and we already use informal markup to indicate quotes. But you could also imagine user-defined tag terms also.
2) No. Or at least, I only think you should consider that as a last resort.
2.5) To me what is most critical in notes is being able to distinguish my commentary from exerpted/quoted content (quotes/blockquote), and to be able to identify from where that content came from (page number, for example; though with tools like Zotero, you could imagine that becomes less relevant, and other ways of specifying location within a document will become more critical).
To me, anything else (emphasis, lists, etc.) would be gravvy I could live without.
3) While I've been thinking about this issue before and it's come up, to be honest, it's not supported in CSL. But obviously my thinking on this is tied to that in part. I'd rather have rules like "if the title within a title is an article, surround with parenthesis, else italicize" rather than to tie this to presentation (which would be really hard I think, and probably unreliable).
This is tricky though. It's weakly supported in existing formats, or not at all. You can support inline (say XHTML) markup in RDF, but it requires changes (rdf:parseType="Literal" on the properties). You cannot support in it MODS, and would have to strip the markup.
In any case, all of this is to say it's one of those big issues that bears further community discussion (maybe with the more technical discussion on the dev list, and user requirements more here).
Foreign words. This is very common in some disciplines, such as mine, history.
Ship names. I'd never have thought of it, but if Chicago says it, it must be so.
<dc:title rdf:parseType="Literal">
<x:div>Some <x:em class="species-name">foo</x:em> and <x:cite>title</x:cite></x:div>
</dc:title>
In that case, a web browser would typically display it correctly without any extra work, and there's enough there to reliably transform it into other output.
Maybe not ideal, but something along those lines.
Johan
help
Some examples:
Smallcaps
http://www.springerlink.com/content/c9654h9700x4nn47/
Non-numeric subscript
http://www.nature.com/nsmb/journal/v2/n1/abs/nsb0195-69.html
Non-numeric superscript
http://biology.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pbio.0020163
The current citation engine in Zotero is being rewritten, and hopefully should be finished in a couple of months. It should come with rich text support for titles. If you don't have the time to wait (I hadn't), you can use HTML tags in the mean-time. It works fine for me. See:
http://forums.zotero.org/discussion/3875/rich-text-in-titles/?Focus=26555#Comment_26555
Is that function available?
I need italics in many title...
Rich text formatting is working in a new CSL processor built for use in Zotero. It hasn't been deployed yet, but if you use the solution that Rintze links to above, your data will very likely carry over smoothly when the Z version with rich text support in the title field arrives.
This is for OpenOffice Writer, but if you know about Wildcards (Microsoft Word) you can probably adapt it accordingly. Don't know enough about the equivalent for WordPerfect.
I use the character "/" in the Zotero item to delimit parts of titles that need to be italicized. This may or may not work for you, depending on whether or not you will use the / character elsewhere just as part of your document. The trick here: find a keyboard character you NEVER WOULD HAVE USED OTHERWISE, like maybe "~".
In OpenOffice Writer:
1) Create a mock text, e.g.,
/This is a test and/ this text is outside of the slashes.
2) Tools-->Macros-->Record Macro (if "Record Macro isn't there, check out this page: http://user.services.openoffice.org/en/forum/viewtopic.php?f=20&t=28782&p=131108&hilit=record#p131108)
3) Ctrl + F to open Find and Replace
4) Click on the "More" button, and check the box next to "Regular expressions"
5) In the "Search for" box, enter "/(.*)/" (without the quotes)
6) Put your cursor in the "Replace with" box, then click "Format." On the "Font" tab, under "Typeface," select "Italic". Click "OK."
7) Click inside the "Replace with" box again, and enter "$1" (without the quotes)
8) Click "Replace All". A confirmation window opens, saying "Search key replaced # times," where # is the number of times the change was made. Click "Ok" to close this window.
9) Uncheck the "Regular expressions" box, then click "Close."
10) Click "Stop Recording" on the Recording Macro window.
11) Give your macro a name you'll remember and click "Save." (I have it under "My Macros." I don't know if it'd be better to put it under "OpenOffice.org Macros" or not; I leave that question to someone more technically advanced than I.)
12) Now, to assign the keyboard shortcut: Go to Tools-->Customize. Select the "Keyboard" tab. In the "Shortcut Keys" window, choose the keyboard shortcut you'd like to use (I use Alt+I), In the bottom half of the window ("Functions"), under "Category," scroll all the way down, and expand "OpenOffice.org Macros." Expand "user," then "Standard," and your macro should be there. Highlight it, and click "Ok>'
13) You should now be good to go. Enter some more mock text, e.g. "/This is another test/ and this text is outside the slashes, hit your keyboard shortcut, and see if it changes the text to italics. If it does, it will also work without a problem with any footnotes or endnotes you've created with Zotero that use the /.../ to delimit the portion of the title that needs to be italicized.
Of course, every time you "refresh" your citations, it'll change the italics back to /.../, and you'll need to hit your keyboard shortcut again. But this'll do the trick.
Hope this helps someone!
All right, in case anyone has tried my solution above (again, this is for OpenOffice Writer), here's an important caveat that I just discovered. Recording the above steps as a macro omits the step of selecting "Italics" in the "Replace" box. In other words, what seems to matter is not whether you select "Italics" for "Replace" while RECORDING the macro; what matters is whether, at the time you EXECUTE the macro, the "Italics" attribute is STILL active for the "Replace" box, were you to hit CTRL + F. If you record the macro, use it successfully several times, and at some point down the road do another find/replace where you remove the Italics attribute from the Replace box, the macro will no longer work (that is to say, it will run, removing your special characters around the text you want italicized, but the text will remain in normal font).
I have tried everything I can think of to fix this (e.g., re-recorded the macro where I SELECTED the Italics attribute, and then DE-selected it with "No format" while still recording), but to no avail. I know very, very little about programming macros, working within the code. If someone who does know how to code macros knows of a solution to this, I for one would appreciate very much to hear it. As things are, I'm going to have to remember to do a mock find/replace, and set the replace box to Italics, before actually running the macro.
A decision analysis of flow management experiments for Columbia River mountain whitefish (<i>Prosopium williamsoni</i>) management
Atlantic salmon <i>Salmo salar</i> L., brown trout <i>Salmo trutta</i> L. and Arctic charr <i>Salvelinus alpinus</i> (L.): a review of aspects of their life histories
Any chances that Zotero will be able to handle this correctly by using italic font for the words in between the tags? From the discussion above I am not sure what the answer is... Thank!.
http://forums.zotero.org/discussion/3875/rich-text-in-titles/?Focus=26555#Comment_26555
Because, honestly, I think it is ridiculous that this problem has not yet been solved.
I love Zotero, and I'm using it actively every day, but this issue has been raised in the pre-beta days and we still depend on ugly workarounds. I know dozens of academics who have given Zotero a try and immediately stopped the moment they realized that there was no support for something as simple as a title within a title. It's extremely off-putting to anyone working in literature (e.g.), and the Zotero-developers seem to realize this.
Honestly, with all the features of version 2, I was sure that this was coming as well - it seems much more important than most other features (e.g. sync) for which easy workarounds existed before. I hate complaining with free products, but I am concerned that this is still a problem.
Just putting on some pressure. This really should have highest priority, and you really should be able to put a date on it. Let me just show you how this looks from a user's perspective:
Apr 8th 2007: "Semantic metadata fields and HTML in notes should be able to happen pretty soon"
Jun 26th 2009: "the weather report says "pretty far along"
Jul 27th 2009: "The current citation engine in Zotero is being rewritten, and hopefully should be finished in a couple of months. It should come with rich text support for titles."
Sep 6th 2009: "The main thing at the moment is to get the existing solution out there so we can get some feedback."
Feb 10th 2010: "Rich text formatting is working in a new CSL processor built for use in Zotero. It hasn't been deployed yet."
It's like a constant promise that isn't being fulfilled (and I know that I've simplified many of the above comments by taking them out of their context).
Is anyone assigned to work on this? Can you give us a hunch of a date?