About this board

    Not editable

    Need of another transcription without 'corrections'

    Jan (talkcontribs)

    While the 'corrected' transcription makes it easier to understand, learn and teach the Fryas language and spelling, a transcription without these changes will be essential for study of spelling and grammatical varieties and (possible) errors. The 'corrected' version may even have to become the secondary option on the wiki, in stead of primary, as it is now. What would be the easiest way to generate a proper transcription in which the recent transcription improvements are included, but not the 'corrections'. Added or removed spaces or dots are another question. I will think further about how to deal with those.

    Pax (talkcontribs)

    It may be possible to create a script that reverses the corrections listed in the transcriber notes, but only if they are written consistently in the format of PREVIOUS → CURRENT. For example, no English descriptions like “added a missing dot.” I will look into this sometime. I prefer the corrected transcription as the default, since it is a more pleasant reading experience. The version only with typist corrections could be provided as a text file, since that version is relevant to a small number of specialists, not general readers.

    Even so, I think the fully corrected version is more useful when creating a grammar. A grammarian finds the most common inflections and conjugations (already done in transcriber corrections) and prescribes the grammar (phonology, morphology, syntax etc.) from that foundation. A student uses that grammar to understand the text and to write new text. The Oera Linda text that the student reads is a corrected version that does not suffer from the many inconsistencies in the pure original. Furthermore, example sentences for instruction would be consistent with the corrected text.

    Otherwise, imagine in a different scenario the frustration of being a student trying to read and understand a Latin text full of inconsistent spellings, and furthermore trying to write anything in Latin not knowing whether to pick this or that spelling because the grammar prefers not to prescribe correct usage when possible.

    This is my opinion of course. Linguistic prescription is more useful to the revival of Fryas than purely linguistic description.

    Jan (talkcontribs)

    I don't think the changes were written consistently enough for such a script to function flawlessly. All pages may have to be checked first. Dots and spaces are of less significance for analysis of the original spelling in the MS.

    Indeed, the more consistent version will be easier to read and use for learning, teaching and standardization. But for research of the MS, e.g. conclusions about varieties within and between texts, an unmanipulated transcription is essential.

    Whether the unmanipulated transcription or the adapted version will be the primary option can be decided later. Perhaps they can become options like the fonts are now.

    SKRIVA varieties/derivates

    Jan (talkcontribs)

    I made an inventory of all SKRIVA-related varieties. Since they are so many, I decided to only correct the most obvious mistakes and rarities, sometimes also based on page/author/idiom consistency.

    Summary: I will align

    • [00a] [147] [163] perfect SKRÉWEN and SKRÉVIN with the clearly more common SKRÉVEN
    • [004] [005] perfect ASKRIVEM and A.SKRÍWEN with ESKRÉVEN (SKRÉVEN is the regular form but there are 4 cases of ESKRÉVEN which I will keep for now)
    • '[103] SKRIV.FILT with 5 x SKRIF.FILT
    • [076] [116] SKRÍVER and SKRIVER with 4x -WER (Í and I variety was kept)
    • [087] infinitive BISKRIVE with -A
    • [068] SKRÍWEN (seemingly perfect) should be infinitive SKRÍWA/SKRIVA; latter variety was chosen to have on-page consistency

    Numbers spelt with both words and digits

    Pax (talkcontribs)

    I think it would be useful to readers to write the numbers in brackets. For example: TVÉLF.HVNDRED. SEX ÀND FIFTIGOSTE [1256]. Same in 00b, and anywhere else where a number is spelt out in words. @Jan

    To compare: in other places in the manuscript, only a number is given, where you have written it out in words. For example, on p. 117: 1888 [ACHTTIAN.HVNDRED. ACHT ÀND ACHTANTICH]

    Jan (talkcontribs)

    Yes good idea, but then in normal font (as the line numebrs) so it stands out better? They should not be confused with line nrs though. Let's think about it.

    general thoughts about transcription

    Jan (talkcontribs)

    VV or W

    Sometimes two V's are clearly not connected as a W, sometimes they are separated, but there will me many cases of doubt. I would suggest we agree to always transcribe VV as W, unless in the theoretical case of a compound word of two words, one ending with a V and the other starting with a V. I don't see the need to note changes of VV into W .

    Tildes ( ~ )

    They are not used consistently in the manuscript. Sometimes to suggest a reading pause or start of new paragraph, sometimes to fill space. Often they are not used where they would be expected. I would suggest to leave them out of our transcription (which currently is in line with our paragraphs in transaltion), unless it is one like Jensma made, following each line as in manuscript. In the theoretical case of them being really functional, they can be included.

    Dots used to separate words or hyphenate at line ending

    Where dots are used to separate compound words, they should be used in transcription, but when their function was separate words that were simply written too closely together (to avoid confusion), they have no function in the transcription. In cases of doubt, these can be noted below the page. Similarly, when used only to break a word at the end of a line, they have no function in the transcription and would only be confusing. Sometimes a clear underscore is distinguishable where otherwise a dot is used and there are many cases of doubt. Thus far I chose to only use dots, for consistency. Also, sometimes words seem separated by a blank space where they should be connected (example [11] WR LANDISK → WRLANDISK). I.m.o. it would be to much to note all these transcriber's choices, as no clear rule can be defined. I would only noe cases of doubt.

    Example: THRJA.THÛSOND. FJVWER HVNDRED ÀND NJUGON ÀND FJVWERTIGOSTE and TVÉLF.HVNDRED. SEX ÀND FIFTIGOSTE: the first dot is used to connect THRJA and THÛSOND (to 3000), the second as a comma, but why is there no dot between FJVWER and HVNDRED (400)? Since words connected by dots (without blank space) are not automatically split when not fitting on a line, this can lead to long words shifting to the next line where we would want to hyphenate and split them. Even when they would be split like that, it would be confusing as they can then be confused as periods (end of sentence). In the above example I would prefer to use blank spaces instead of dots, even though they are used in the manuscript.

    Copyist corrections vs. ink stains

    On the HQ image [11] ?A is clearly RA with an ink stain, whereas [19] NJUG?N is NJUGEN changed into NJUGON. The latter makes sense because NJUGON can have been pronounced as NJUGEN. However, in [11] WRLANDISK, the D will not initially have been Í, as Jensma thought, but rather a slip of the pen. While NJUGON may be interesting to note, the other two are not, imo.

    Copyist errors or missing (torn off) letters, corrected/supplemented by transcriber

    These are most important and should be noted. Examples: [00b/06] SRÉKATH → SPRÉKATH and [022/07] H???A → HELPA.

    A or Á and A or À

    These are sometimes difficult to distinguish. In case of doubt, it's best to choose A. When an accented letter seems to be used, wheres the same word is elsewhere always spelled without it (or vice versa), or it really seems to be an error, it can be dealt with as a transcriber's correction.

    Pax (talkcontribs)

    OK, I will keep this in mind.

    I will ignore tildes then, since they only seem to serve the same function our modern horizontal rule does.

    Your observations on the dots make sense and correspond to most of what I noticed. There are indeed sometimes confusing cases. On numbers, to make sure I understood: do you suggest e.g. THRJA.THÛSOND → THRJA THÛSOND and TVÉLF.HVNDRED → TVÉLF HVNDRED? That is, transcribe them as if they were written like English three thousand and twelve hundred?

    There are some cases where what looks like A is transcribed as Á or À, often in words like LAND/LÁND or THAT/THÀT. I have noted those in my personal notes so far.

    Jan (talkcontribs)

    I am not certain, but with the numbers THRJA.THÛSOND, FJVWER.HVNDRED and TVÉLF.HVNDRED, the dot is functional: 3 x 1000, 4 x 100, 12 x 100.

    I remember that I may have added accents when I thought that served consitency, for example when in the same text the word was mostly spelled that way. Perhaps I would do different now. Too much time has passed since I made and later changed the transcription. I did not keep good notes of it.

    manuscript pages

    Jan (talkcontribs)

    The HQ images provided by Tresoar and used since the 3rd printed edition are* 2777 x 3686 pixels; print size between 23.53 - 23.99 by 31.24 - 31.85 cm (between 9.27 - 9.45 by 12.30 - 12.54 inches); resolution between 15.75 - 118.00 px/cm (294.00 - 299.72 px/in). [*based on sample of 10 random pages, opened in]

    File size per image is between 4.3 and 5.6 MB. Total of all 190 pages: ca. 0.9 GB (according to windows explorer).

    Ideally, new photos/scans should be made (on a white background and in a higher resolution?), with the lines all straight. I know too little about (use and printing of) digital images to have a good opinion.

    Line numbers (1, 5, 10 etc.) have been added manually.

    For now, we could add the HQ images available as an option. Adding them on all transcript pages may increase their loading time. I can't judge if they would take too much space on the server.

    Pax (talkcontribs)

    I am not an administrator, but I would assume there is enough space for the images. It should be OK to use the HQ versions on the wiki, since MediaWiki generates thumbnails that take up less bandwidth. On each manuscript page, one would display a downscaled thumbnail that is still big enough to read. Users can then click it to download the original HQ image if they need it.

    Jan (talkcontribs)

    Can I send the files to you, so you can upload them? I succeeded last time but found it complicated and don't remember how I did it. I'm afraid I would mess things up.

    Jan (talkcontribs)

    I suggest (and would prefer*) to prioritise correction of (and in case of 2 adding notes):

    1. obvious errors by me, for example [012] LIK VNBIKÀNNADE → LIK EN VNBIKÀNNADE

    2. words corrected/supplemented by transcriber, for example [00b/06] SRÉKATH → SPRÉKATH and [022/07] H???A → HELPA.

    (* it would currently take too much of my time to review all changes and adding of notes on the wiki)

    Pax (talkcontribs)

    OK, I will focus on that. Other minor typographical differences can be noted another time in a second proofreading.

    Jan (talkcontribs)

    Great. Alternatively, if you notice them anyway, you might keep them in file for now, not add them to the wiki yet. But I suppose it would take much more of your time.

    bug in discussion function

    Jan (talkcontribs)

    There is/are still (a) bug(s) in the discussion function, in particular related to switching between visual/source editing. I have lost much work (and patience) trying to use it, whereas this function is really important. Can it please be tested and solved, Henk or Pax? Also, with the separate topics it's very different from how it's organised in Wikipedia. I think the way of Wikipedia is more clear and easier to use.

    Mf (talkcontribs)

    Hi, For the discussion area we are using an extension called 'Structured Discussions' also called 'Flow', to facilitate separating topics. Of course it is possible to disable this function, but I'm not sure what effect that would have on the existing topics, but I'm quite sure it will be a lot of work restoring everything... Possibly the information can be exported and than re-imported when the extension is disabled. I would have to investigate...

    Jan (talkcontribs)

    Thanks for the quick reply, Mf. If the bug of visual/source-editing can be fixed, we then better keep the current separate-topics structure.

    Mf (talkcontribs)

    Possibly, today's update fixed it. But if you keep having issues, please let me know, preferably by mail, what the bug involves and I'll see if something can be done...

    There are no older topics