Skip to content

Reverse CSS


Jakob Voß and I have been looking at how we can usefully extend the Scholarly Copying mechanism (very similar to what you built as Augmented Copying and I had in Flow as Copy/As Citation), using Visual-Meta.


The reason for ‘visual’ meta is for archival purposes, ease or adding and extracting and what Jakob termed ‘Reverse CSS’, which I found perfect: The point of authored documents in Liquid | Author is that they should contain semantically meaningful visual presentations, which is why headings are always tagged as headings and so on.

Citing Multiple Contributors

The specific issue we are looking at now is what would happen to a book like The Future of Text, which will have multiple contributors–how can the system easily know and put into the clipboard what contributor’s text was selected and copied?

It would be relatively easy to make up some schema for this where all the data is hidden but I suspect it would be very hard to get it implemented by different software developers. Therefore we should look into describing what formatting is used to denote the start of a section authored by someone other than the editor, how that person’s name or names is likely to appear and what the end of such a section might be. Also, there should be a list of contributing authors and their associated OrchID and other identifying markers. Therefore (this is of course not strict BibTeX formatting, it’s laid out for logic and dialog):

contributing author section = {Heading level 1}, author name style ={bold}, end = {next heading},
contributing authors = {Douglas Carl Engelbart, Ted Nelson},
contributing authors orchid = {…, …},

How does this look? Also, would this particular case work better with embedded meta and if so, would that be hard to create?

Further issues include how a reader might copy a citation or even a citation by a contributing author. It should be possible to pasta with citation information including ‘as quoted in’ and so on.

Further Issues

Time data should be included, such as a time in a video or a time in the real world, which can then be used for further searches/views, including frames if video:
time = {33mins, 20 seconds, 12 frames}

Real-World Exact Time:
date = {16/5/2013, 33mins, 20 seconds, 9 milliseconds} – where the resolution can go as far as desired.

Support for retrieved by dates:
urldate = {16/5/2013}

Formatting should be in the way CSS styles are formatted so that if colour is used, it can be mentioned as well.


But my main question here, what is the best way to deal with contributing authors?

Published inCitation MetaScholarly CopyVisual-Meta


  1. So the only one major decision here is if you’re making visual, inline metadata as part of the visible document;

    You can either embedded metadata in sections, or put it all at the end and reference the sections somehow. I’m still very unconvinced by “visual meta” as a solution. For this particular case I would, in HTML, use tags

    Marking up PDF is almost a nonsense concept, as it’s a presentation format not a markup format. Without doing something clever under the hood there would be no way to know which section(s) a copied range of information came from.

    • I agree with how to deal with HTML but for this I have to deal with PDF since that is where the academic world currently ‘lives’

      Can we not put into the visual-meta that new sections are marked thus and author names are marked thus and end when there is something other than body text, for example?

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.