missing metadata: e.g. for leveling, antiqua, special characters, letter spaced, umlaut, old greek

[tboenig]

@bertsky: I am missing some metadata for the following cases:

• the levels of ground truth
• antiqua font with or without "ſ", with "ß" or "ſs" or "ss" (e.g. https://www.deutschestextarchiv.de/book/view/goethe_metamorphose_1790?p=10)
• legal/technical/commercial/linguistic/... Special characters
• letter spaced and/or bold print
• umlaut äöü and/or aͤoͤuͤ
• old greek (greek diacritics...)

Can you add something? @bertsky

[bertsky]

• the levels of ground truth

note: there is already a top-level processingLevel – but I don't understand it TBH (an it's not used AFAICS)

• antiqua font with or without "ſ", with "ß" or "ſs" or "ss" (e.g. https://www.deutschestextarchiv.de/book/view/goethe_metamorphose_1790?p=10)

perhaps other systematic conventions in the writing system, like whether or not ꝛ is present, umlauts (see below) etc

• legal/technical/commercial/linguistic/... Special characters

I don't know if these fit, but there are already domains under topic/...

• letter spaced and/or bold print

suggestion:

• data-attributes/document-related/visual/text/font/bold-face
• data-attributes/document-related/visual/text/font/letter-spaced

• umlaut äöü and/or aͤoͤuͤ

question: is it relevant to single out the rare case where both are present?

• old greek (greek diacritics...)

esp. whether polytonic or monotonic or old/pre-standard

All these character set-specific distinctions probably don't fit well into a flat tagging system...

Can you add something?

Perhaps under dataTransformation/... one should have tags for

• binarization (whether the image has been converted to bitonal during digitization)
• compression (whether the image has undergone lossy compression during digitization)

(I would not subsume these under enhancement.)

[tboenig]

thank`s @bertsky

The GT labeling metadata complements the existing metadata that is present in each Page file. The requirement is that the GT is in PAGE format. The page format provides for a large amount of metadata. Especially in the Text Style areas specific metadata can be assigned. TextStyle: https://ocr-d.de/en/gt-guidelines/pagexml/pagecontent_xsd_Complex_Type_pc_TextRegionType.html#TextRegionType_TextStyle

to 1.

• good idea, in my opinion this can be included.

to 2. In Labeling, a distinction is only between Antiqua and Blackletter. Typographic forms or spellings should regulate the levels. For example the text is in Antiqua see example: text is without long s, labeling: level1, antiqua text is with long s, labeling: level2, antiqua

to 3. legal/technical/commercial/linguistic/... Special characters

• There are level rules for this. See: https://ocr-d.de/en/gt-guidelines/trans/trGrundsaetze.html

to 4. Font with blocking and/or bold print This is to be implemented in Page. See https://ocr-d.de/en/gt-guidelines/pagexml/pagecontent_xsd_Complex_Type_pc_TextRegionType.html#TextRegionType_TextStyle

to 5. Umlauts: umlauts as äöü and/or aͤoͤuͤ

• Proceed similarly to the S-forms. See point two.
• Level 1: äöü, Level 1, blackletter
• Level 2: aͤoͤuͤ, Level 2, blackletter

to 6. Special orthographies and changes of spellings which were/are officially enacted (e.g. spelling reform in German, monotonic orthography,...).

• There are level rules for this.
• The specification of the language takes place in the page file.