These are very interesting examples.  Rather than distinguishing between
them one might recognize what they have in common: they are marginalia,
put in the same field of vision by the author or printer, who does not
care about the precision of the visual link as long as it can be
"eyeballed" by the reader. The most precise way of rendering the
relationship between text and marginal gloss would be to recognize its
fundamental lack of precision. The  marginal note says something like "I'm
close enough to let the reader figure out why am I here and what I am
doing."  Should the encoding try to do more?

In that case <note place="margin"> would work for all of them, and
wrapping the dates in a date element would tell the machine what any human
reader would know or figure out.

The subject of this thread, combined with picture, would make a nice
addition to the Guidelines, but the Guidelines actually do quite a nice
job explaining marginal notes, although the example from the Ancient
Mariner means more to a reader who knows the poem and knows that Coleridge
is playing with a genre of marginal annotation that by his time was

The historian of print will note that marginal glosses have come back,e.g.
in the the Landmark Thucydides, perhaps under the influence of the Web.

On 8/31/11 7:26 PM, "Paul F. Schaffner" <[log in to unmask]> wrote:

>On Wed, 31 Aug 2011, Lou Burnard wrote:
>> I can see the attraction of using <label> rather than <note> where the
>> function of the marginal note seems to be to label something rather
>> comment on it (though it's not a distinction I'd like to defend in a
>>court of 
>> law). I cannot see any justification for using <argument> for these
>> though: an argument is supposed to summarise the whole of something,
>>and be 
>> prefixed to it, like a table of contents.
>I didn't understand how Martin was using <argument> either, to be
>honest. Perhaps as suggested below under example 5? (i.e., where
>we essentially *construct* an argument out of marginally dismembered
>> Paul's milestones seem to me to be verging on the pathological.
>Thank you, Lou. This is a comment to treasure!
>Martin I am sure has his own examples, but here are a few
>to clarify our (admittedly muddled) approach to marginalia.
>Examples always help.
>Our approach is (again admittedly) influenced in part by the
>inability of our display system to render notes in a
>reasonable way, perhaps biasing us against use of note in
>marginal cases, so to speak, when other options are available.
>For me the decision tree always forks first at the question:
>is this thing in the margin best regarded as intrinsic or
>extrinsic to the main text. If the latter, it is probably
>a note (or milestone). If the former, one might wish to move it
>formally *into* the text by tagging it as head, label, or even
>Here are some examples, beginning with the pathological ones, and
>ending, I think, with things more like Martin is actually asking about.
>The marginal numbers here I would call extrinsic, marking
>divisions in the text and indicating the text's progress
>according to a standard reference system. We would definitely tag
>these as <milestone>s (probably with @unit="verse"). They
>represent progress through a very standard reference
>system, indeed, viz., the verses of Psalm 90.
>These we would also tag as <milestone>s (probably with
>@unit="year"). They represent the text's progress through
>the standard reference system of the calendar. (BTW, they
>are *not* line or stanza numbers.)
>We tend to leave a lot of these as <note>s, but in this case
>I would feel free to interpret them as intrinsic to the main
>text and tag them as labelling the paragraphs against which
>they lie.
>   <p><label>Of the Asse</label> The Asse denotes a good Servant
>   of Slave, that is profitable to his Master; it signifies also a
>   foolish and ignorant person.</p>
>   <p><label>Of the Mule.</label> The Mule signifies malice and
>   foolish imaginations. Artemidorus saith it signifieth sickness
>   to him that dreams he saw one.</p>
>This is much like the previous example, and again I would feel
>free to interpret the marginalia as somehow heading the accompanying
>text, rather than standing outside it and commenting on it, and
>therefore of being tagged as <label> (if the accompanying text is tagged
>as <p>s); or as <head> (if the accompanying text is tagged as many small
>    <div><head>1. Full Armour.</head>
>    <p>ALL Souldiers comming to their Collours to watch or be
>    exercised, shall come fully Armed upon paine of severe correction.</p>
>    or
>    <p><label>1. Full Armour.</label> ALL Souldiers comming ...</p>
>This example illustrates two different kinds of marginalia.
>"Section 19." we chose to regard as a div heading. I.e.,
>we began a new div @type="section" n="19" at the beginning
>of that paragraph, and placed <head>Section 19.</head>
>at its head. If we had *not* chosen to insert the div
>break there, our free interpretation of <milestone>
>would have encouraged, or at least allowed, us to tag it
>as <milestone unit="section" n="19.">
>But the remainder of the marginalia on the page is of a
>different sort. It constitutes in toto a running
>summary of the argument, an outline complete with
>numbered points and subpoints. And breaking it up
>into separate <note>s does it a disservice (unless
>you use TEI mechanisms to string the notes together),
>since otherwise you are left with individual notes
>that are meaningless
>   <note place="marg">And by renewing vs.</note>
>In cases like this we sometimes -- very rarely --
>choose, in violence to the source, to gather all the
>related marginalia for a given section into a single
>continuous <argument> and place it at the beginning
>of the div.
>  <argument><p> 1 ... 2 God's spirit shewes it selfe
>  an holy Spirit, in begeting good motions. And by
>  renewing vs. ... 3 ... </p></argument>
>This example also has two different kinds of marginalia.
>"Deut. 28." and "1 Cor. 11.29." are standard bibliographic
>notes, tagged as <note place="marg">.
>"Sicknesse from God." and "Sicknesse through sin." are marginal
>summaries, similar in some ways to examples 3, 4, and 5. They
>stand on their own, so they certainly do not require the radical
><argument> treatment.In truth, we would probably leave these as notes,
>placed at the beginning of their respective paragraphs. If
>we were ambitious we might try to distinguish them from the
>other notes by @typing them distinctively.
>But they are also close enough to the examples that
>we chose to regard as <label>s intrinsic to the
>main text that we could rationalize tagging these as
><label>s too, in effect moving them into the text
>and out of the margin.
>  <p><label>Sicknesse from God.</label> Whether then sicknesse
>  doe fasten on vs, through an externall bad ayre, or some
>  internall distemper of blood ... </p>