pfefferniels / probstuecke-digital Goto Github PK

The facsimile section for the first edition shows wrong pages on opening.

The icons in Mirador viewer are broken/missing, making it hard to handle.

Special characters like the long s (ſ) aren't rendered at all in the exported PDFs. And (which is even worse) – all raised sixths (6⃥) come out as plain 6ths. We should investigate, why this is happening …

It would be nice to have an option to normalize text (umlauts and long s).

Also the Staves and Clefs option should only appear, when there are actually staves and clefs (i.e. not on the front page).

Trying to "view occurrences" in the bibliography throws this:

failed fetching references SyntaxError: The string did not match the expected pattern

Any idea what's happening here?

As the svg rectangles in layer elements overlay the staff lines (e.g. see http://probstuecke-digital.de/7) it may be good to use pure css to highlight something.
We could introduce something like:

g.blink {
animation: blinking 1s infinite;
}

@keyframes blinking {
  0% {fill: red;}
  50% {fill: black;}
  100% {fill: red;}
}

and use js to change the class of all referenced elements on click.
The continued blinking helps to better track down the highlighted elements.

In some pieces (e.g. PS11) there are annotations by Mattheson in the left hand. These too shouldn't be displayed by default in the edition. That can be achieved by XSLT easily. Question is: how to encode it in a good way?

The website shows the favicon from React.
Maybe we want to exchange it with something better?
I tried to style the "O" from "Organisten-Probe" from the title page:

We should add a page that states:

1. the goal of the edition
2. How to use it
3. the edition guidelines
4. the persons involved

How should we encode figured bass (in particular stacks of figures) in TEI? The question of how time signatures can be encoded seems related to me.

It would be nice, if the headers would be brought in line and enriched with more information.

We should have a 'real' title page, stating the name of the project etc.

Also, the regularized text I added to the front page is shown normally, but should be a tooltip or pop-over ideally …
For now I would simply make:

tei-choice tei-reg {
  display: none;
}

I'd like to propose using <pc force="weak">-</pc> for the hyphens at line breaks (as suggested by the TEI guidelines). This could be added by a simple regex on all the TEI documents. I don't feel very good about the regex replacements by JS everytime a Probstück is loaded …
This would allow us to simply hide linebreaks by CSS: tei-pc[@force='weak'] { display: none } and tei-lb { display: none }

https://probstuecke-digital.de is currently not working.

Facsimiles do not work on highlighted bars.

Opening the characteristics on key and meter is a bit tricky. Once opened, you have to click again on the key / meter to close it. Would be more convenient to close it by clicking somewhere outside (Bootstrap's dismissible popovers).

Trying to open it again then fails.

Look here and add a LICENSE.txt

Names with references to the GND shouldn't link directly. Better would be a pop-over showing the normalized name from the GND with a separate link.

Following some Prob-Stücke there is a remark by Mattheson, especially that one below PS8. How to deal with those? Where should they be encoded and how?
(In the PS8 case we would then have two key-overlays)

For now it is enough to say something like "The encoding is based on the DTA-Basisformat", but in the long run, we should provide an ODD, not only to make sure that all files apply to our own rules.

Ist das der "way to go"? Ich denke, der Erkenntnisgewinn ist deutlich größer als einfach stumpf Erst- und Zweitausgabe separat zu kodieren. Ist aber zugleich ziemlich aufwendig ...

While in DTA-BF <pb> marks the beginning of a page, <lb> is used at the end of a line.
The TEI Guidelines now make it very clear that it should be used to mark the beginning of a line. We should follow the TEI Guidelines.

See:
https://www.tei-c.org/release/doc/tei-p5-doc/en/html/ref-pb.html
https://www.tei-c.org/release/doc/tei-p5-doc/en/html/ref-lb.html
and
deutschestextarchiv/dtabf#33

(can't reopen #72)

Instead of encoding them manually, all facsimile_*.json files should be generated by an npm script that walks through all MEI and TEI files, extracts the information from the <facsimile> elements and compiles them into the JSON files.

Currently, all facsimile files are stored on the public/facsimile folder. All references between facsimiles images and the transcriptions are stored in the annotations_{1st|de}.tei file.

It would be good to structure the facsimile data in a better way:

1. MEI files contain their own facsimile references as music is concerned
2. TEI files contain their own facsimile references as text is concerned
3. Additionally, an AnnotationList service is provided that keeps the combined data of (1) and (2) as IIIF annotations

An annotation on a whole page of music might look like this:

    {
      "@context": "http://iiif.io/api/presentation/2/context.json",
      "@id": "https://probstuecke-digital.de/iiif/1/annotation/facsimile_de-p386",
      "@type": "oa:Annotation",
      "motivation": "sc:painting",
      "resource": {
        "@id": "http://probstuecke-digital.de/iiif/1/score.mei#xpointer(//pb[n='386'])",
        "@type": "dctypes:Image",
        "format": "application/mei+xml"
      },
      "on": "https://api.digitale-sammlungen.de/iiif/presentation/v2/bsb10598495/canvas/386"
    },

Or an annotation on a measure of music:

    {
       "@context": "http://iiif.io/api/presentation/2/context.json",
       "@id": "https://probstuecke-digital.de/iiif/1/annotation/facsimile_de-measure1",
       "@type": "oa:Annotation",
       "motivation": "sc:painting",
       "resource": {
         "@id": "http://probstuecke-digital.de/iiif/1/score.mei#xpointer(//staff[xml:id='m-40'])",
         "@type": "dctypes:Image",
         "format": "application/mei+xml"
       },
       "on": "https://api.digitale-sammlungen.de/iiif/presentation/v2/bsb10598495/canvas/386#xywh=175,134,439,185"
     }

And of course annotations on the text ...

Any comments/suggestions? @rettinghaus ?

For some files MEI3 is use, others are in MEI4. If there isn't a good reason for this, all files should use the same version.

For some reason, after a <hi> element there is an extra-space inserted in HTML view., messing up punctuation. Any idea why this happens? Is this a CETEIcean bug?

Instead of splitting references up into different pointers, like in Prob-Stück 10:

<ref target="m-35">Die</ref> <ref target="m-57">erste</ref>
<ref target="m-80">Zeile</ref> <ref target="m-105">dieses</ref>
<ref target="m-128">Stückes</ref> spielet man

we could use <annot> in the MEI to collect all members with a @plist and point directly to the annotation, like:

<annot xml:id="annot-01" @plist="#m-35 #m-57 #m-80 #m-105 #m128" @label="Mattheson's first remark"></annot>

Music examples in the text shouldn't be forced to the width, but instead rendered according to the width like in the MEI Viewer:
https://www.verovio.org/mei-viewer.xhtml

Recently there popped up some errors, e.g.:

• editorial notes aren't rendered correctly (i.e. not at all)
• double bar line / end bar lines are not shown
• musical examples are (always) shown with modern clefs

Some MEI files still have a GPL license remark. Should we move all of them to the Creative Commons Attribution 4.0 International License, as most of the our XML files are?

The facsimiles are too small, the size cannot be changed (tested Safari, Chrome, Firefox).
I propose to add the IIIF bitonal option to the quality selection list.

The <foreign> element should be used, but not to denote a different font in the source.
Instead a <hi> element should used in addition: <hi><foreign>Nota</foreign></hi>

[see title]

Similiar to #6 the headers of the TEI documents should provide more information

In some pieces the displayed key signature disagrees with the tonality from a modern perspective.
On modernizing the clefs, shouldn't we also modernize the key signatures?

Bernhard, Bottari, Martialis, Morhof and others(?) don't reveal their appearances in the Index of Persons.

What to do with measure numbers in examples? If they were just extracts from the whole piece we could leave the original numbers in there. But usually they're more than that. Do they deserve a local independent numbering or should we just drop them entirely?
For pointing to measures in the score we use @corresp. From a technical perspective way to go. But it is not for humans. Should we give a @label with the corresponding measure number?

The Si suona etc. is no longer shown.

While generally text normalization using DTA's CAB service works fine, it introduces some mistakes, e.g. Schleiffer gets normalized to "Schluffer" and some more. Probably we should collect those and correct them after normalization again.

As discussed some time ago, I'd like to start moving the MEI/TEI encodings into a separate repository, that we can include as a git submodule. Are there any objections, hints, comments?

Bootstrap resets non-serif fonts system-specific differently, making some glyphs we use in the edition look ugly. We should look for a good font we can use at least for the comments.

This one looks promising, but should be tested thoroughly: https://fonts.google.com/specimen/Scada

In the very long text on (or better: following) the 24th Probstück Mattheson reflects, just like in the "Vorbereitung zur Organisten-Probe", about all sort of things (temperature, notation problems etc.), but says nearly nothing about the Probstück itself. To which extent should we or shouldn't we reproduce this text?