There are a lot of questions about where one aspect of ‘text’ ends and another begins. The <header> and <text> distinction of a TEI document has its roots (sort of) in the relationship of a “book” to its record in a catalog. header:text = card: book. That model has been extended to the point at which it may no longer work. A lot of stuff is now put into headers that one would never have put into a library catalogue. And the <teiHeader> now has <xenoData> into which you can put anything as long as it isn’t TEI.
I am part of a project that aims at “re-mediating” the heritage of Early Modern English print, using TCP texts as a point of departure. Making the texts more readable is one goal. Making them more computationally tractable is another. For that purposes up-to-date versions of Morgenstern’s Spectacles are helpful(https://scalablereading.northwestern.edu/?p=229 ). They create various abstractions. Some of them can be produced on the fly (e.g. lists of words with counts), and it is not worth storing them separately. Other depend on complex comparisons of a text with other data. Those comparisons may be computationally expensive and serve as inputs for further operations. So it may be worthwhile keeping them as “supplementary information packages” of one sort or another. The relationship of ‘data’ and ‘metadata’ becomes fuzzier the more closely you look at it. It may be metadata all the way down. For something like Early Print the “real” thing may be half a million hooks (or catalogue numbers) in the English Short Title Catalogue. On each of these hooks you can hang a lot of different things, among them a text that still can claim with some justice to be the source of all the other things that hang from the same hook.
As for discursive statements about editorial policy, Dario Kampkaspar and Martin Holmes offer good advice. The header almost certainly does not offer an inviting space for that genre. Digital texts of the future will increasingly be sets of “related items” with multiple and overlapping hierarchies that will challenge the “either or” of <header> and <text>.
Thanks very much, Sam. That is exactly the kind of experience I was curious about. When pondering the arguments, however, I start wondering if we are still on the right track when we attempt to describe e.g. our encoding parctice and use of markup in <front> (presumably nobody would, but it is the kind of information relevant for the digital edition). In other words, what sort of introduction is really appropriate for native "digital" editions and do we have to stick to the "usual" way that comes from the analog? The other point I am still concerned about is the obvious duplication of information. For instance, the <editorialDecl> contains markup for <correction>, <normalization> etc. This is exactly the kind of information I would expect in a critical introduction. Of course, we could encode it in both the <header> and <front>, but practice shows that editors tend to neglect the header in favor of the free introductory text. Sure, <front> or <body><div> provide much more flexibility and perhaps I wouldn't be able to write the introduction the way l was used to when trying to populate the header only , but on the other hand the header is much more precise in its markup and <front> isn't a dedicated place for editorial material and could also be used for a <front>-part of the source itself ( belonging to <text>; so it is ambigious for a computer and requires additional explanation or customization). This seems to be a disadvantage to me.
Am 29.10.2018 um 16:40 schrieb Huskey, Samuel J.:
When we were developing the encoding guidelines for the Digital Latin Library's Library of Digital Latin Texts, I attempted to convert the preface for an edition of a classical Latin text to the TEI header. That experiment made it clear that the TEI header is great for metadata about the publication and other factual details, but it isn't ideal for articulating the editor's argument(s) and method(s), which are integral to a critical edition and should be presented with the text. That's why Hugh and I decided to use <front> for the prefatory materials (https://digitallatin.github.io/guidelines/LDLT-Guidelines.html#front-matter).
From: TEI (Text Encoding Initiative) public discussion list <[log in to unmask]> on behalf of Thomas Stäcker <[log in to unmask]>
Sent: Monday, October 29, 2018 9:23:21 AM
To: [log in to unmask]
Subject: Re: critical introduction
Being a trained classicist myself I confess that I felt somehow troubled, too, but on the other hand I am more and more convinced that we have to treat that kind of introductions differently . Yet as the proof of the pudding is in the eating, I would attempt to convert a typical critical introduction to the teiHeader to provide a suitable example for further discussion. This might take some time.
Am 29.10.2018 um 14:35 schrieb Martin Mueller:I agree with Hugh. Since we were both trained as classicists, this may be a hold-over from a world of print editions with extended and reflective remarks by an editor. That said, some things are worth holding over, and the proper place for a discursive and reflective introduction is <front>. On the other hand, a digital edition may include a lot of machine-readable metadata for which there are no equivalents in the print world. It may be a nice question whether the teiHeader is always the best place for such data.
On 10/29/18, 7:35 AM, "TEI (Text Encoding Initiative) public discussion list on behalf of Hugh Cayless" <[log in to unmask] on behalf of [log in to unmask]> wrote:
In haste, sitting in a dentist’s chair. My own tendency is almost the opposite. I view the introductory material as a first-class part of the edition, and so would tend to put it in <front>.All the best,HughSent from my phone.> On Oct 29, 2018, at 08:07, Thomas Stäcker <[log in to unmask]> wrote:>> Roberto, Patrick,> many thanks for your feedback. The argument of Patrick is exactly what I had in mind. I actually tend to abolish the <body><div type="introduction"> or <front> solution within <text> and move all the metadata to the header (in the case of a born digital edition), even if there is lot of plain text that extends what usually is understood by the term metadata. My feeling is, however, that there should be a sort of recommendation in the Guidlines addressing this issue, if there is broad consenus about it. At least I am convinced that editorial metadata (what is annother expression for critical introduction) would be much better in future when we encourage editors to use the elaborated vocabulary available for the teiHeader instead of putting pertinent data to <text>.> Best,> Thomas>>>> Am 29.10.2018 um 10:31 schrieb Roberto Rosselli Del Turco:>> Hello Patrick,>> you don't have to convince *me*, since I fully agree with you (I would still consider <front> instead of the <teiHeader> as a legitimate place where to save your introduction to the edited text) on the encoding strategy.>>>> This sort of decision typically arises with projects which have had a Web site to present the work being done for all the project's length, and feel it would be a good place to offer an introduction and/or other appropriate paratext. This is also the case of a web site offering distinct functionality when compared to the edition (the CPD site has a collaborative feature, a list of signa tabellionis for each notary, a timeline of events) which reinforces the idea of it actually being the "hub" of the digital edition, with the latter reachable as a separate location.>>>> Best,>>>> R>>>> Il 29/10/18 10:19, Patrick Sahle ha scritto:>>>> Am 29.10.2018 um 08:49 schrieb Roberto Rosselli Del Turco:>>>> Dear Thomas,>>>> the first thing to decide is *where* you are going to publish the introduction for your digital edition,>>> I find this argument quite surprising as I feel that more and more people see "the data" as "the edition" which is conceptually independent from arbitrary acts of publication (see my theory on transmedialization; and yes, I know about the limitations of such a claim).>>> With this feeling I totally understand Thomas' question which sheds some light on deeper ontological questions on electronic texts. For a spontaneous reaction I would say that everything depends on your own understanding of text, text representation and critical editing.>>> If you emphasize that a critical introduction is describing a text that is containes in <text>, the header seems the obvious place to put that description>>> If you emphasize that the digital representation together with the critical introduction to it are an intellectual work in it's own right and autonomy, the body may contain both>>> But of course we are talking of two different texts here: a represented text that is talked about and the edition as another expression of a text>>>>>> All known, I know. But I wonder whether there is broad consensus on this or still room for discussion?>>> Best wishes anyway, Patrick>>>>>>> because some projects prefer to have a dedicated web site which is actually separate from the actual published edition; in the which case, you can have a minimal TEI header since everything is published elsewhere. See for instance the Codice Pelavicino Digitale:>>>>>>>>>>>> Note that in the digital edition proper you have *part* of a critical introduction, since the content of the <msDesc> element is available via the MS Desc button:>>>>>>>>>>>> (very slow web server at the moment, please be patient).>>>>>>>> We're considering supporting a full introduction to a digital edition in EVT directly inside the latter, however, by expanding the "Project Info" window. What would you think of that? Many scholars prefer a separate web site, but I'm starting to lean towards the "everything in the digital edition" solution.>>>>>>>> Last note, if not in the header I would put such an introduction in the <front> element.>>>>>>>> Best regards,>>>>>>>> R>>>>>>>> Il 29/10/18 08:32, Thomas Stäcker ha scritto:>>>>> Dear list,>>>>> even though I have been working with the TEI for a couple of years I am still uncertain about the function and role of the teiHeader versus a critical introduction that as a rule precedes every serious critical edition. I checked the archive of the list, but couldn‘t find any discussion or recommendation about that matter, but might have overlooked it. Currently we include the critical introduction in the body of the edition (<body><div type=„introduction“>). This solution is taken from examples of analog editions that are encoded according to the TEI and where it is evident that all parts of the text including the editorial introduction have to be represented in the digital version of it either. However, when we turn to a born digital edition things are less clear as we can populate the teiHeader with that kind of meta-information. In addition, the teiHeader offers markup that is especially designed and dedicated to describe that particular kind of information whereas the critical introduction in the body would be less normative. Accordingly, the critical introduction can be regarded as in a way redundant to the Header or even superfluous. My question is if there exists any recommondation about how a critical introduction shall be treated in born digital editions. I feel a teiHeader should be there anyway, but how is it to be distinguished from the critical introduction or how can this relation of the teiHeader to the critical introduction be defined? I'd be most grateful for any hint or recommendation.>>>>> Best>>>>> Thomas>>>>>>>>>>>>>>>>>>> -->>> Apl. Prof. Dr. Patrick Sahle>>> Universität zu Köln <http://www.uni-koeln.de> - Cologne Center for eHumanities (CCeH) <http://www.cceh.uni-koeln.de/>>>> Koordinierungsstelle Digital Humanities der Nordrhein-Westfälischen Akademie der Wissenschaften <http://cceh.uni-koeln.de/DH-AWK/>>>> Data Center for the Humanities <http://www.dch.uni-koeln.de/>>>> Institut für Dokumentologie und Editorik <http://www.i-d-e.de>>>> @patrick_sahle <https://twitter.com/patrick_sahle> - @CCeHum <https://twitter.com/CCeHum> - @ideinfo <https://twitter.com/ideinfo> - @dhd2018 <https://twitter.com/dhd2018>>>>>>> --> ***************************************> Prof. Dr. Thomas Stäcker> Direktor der> Universitäts- und Landesbibliothek Darmstadt> Magdalenenstr. 8> 64289 Darmstadt> +49 (0)6151 16-76200
--***************************************Prof. Dr. Thomas StäckerDirektor derUniversitäts- und Landesbibliothek DarmstadtMagdalenenstr. 864289 Darmstadt+49 (0)6151 16-76200
Prof. Dr. Thomas Stäcker
Universitäts- und Landesbibliothek Darmstadt
+49 (0)6151 16-76200