Expand corpus of attested samples. #24
Comments
|
If you want a full, thorough list of every single tiny bit of Tengwar attested, here is the list: http://www.forodrim.org/daeron/mdtci.html It just doesn't have the actual texts reproduced. The problem is, I am not certain of the legality of doing such a thing. The Tolkien Estate already has some large, beautiful documents available here: https://www.tolkienestate.com/en/painting/calligraphy/calligraphies.html |
|
It should be okay to reproduce excerpts of the original texts as test cases and provide academic references, as denoted by the Forordim reference list. |
|
Okay! Then I think I can do this. I did something similar by copying all of the tengwar texts I could find into my notebook, so this would basically be a digital version of it, right? |
|
@dreamingfifi Yes, and consequently quite valuable to the community. Arno and I previously discussed creating a shared bank of tests. This would form the basis of that. |
|
Feel free to change the notation as well, or request changes. It is not yet particularly ergonomic but is mutable. |
|
For general use English - there are actually remarkably few texts. Tolkien preferred the "Northern Variety" for writing English, which is a full mode. There is so little of General Use English that we use some of the ideas from the full mode to fill in gaps in our knowledge. But, I can gather them up as images, then write them out with a Dan Smith font (should I use Parmaite or Annatar?) |
|
It would be ideal to write the attested examples in a notation like we have
in the existing tests.
It would be possible to create test cases using the Dan Smith notation for
one font or another, but they would not be readable as code.
If you do use Dan Smith bindings, we might be able to create a tool that
reverses them into tengwar and tehtar name notation.
I’m sure you’re aware that Parmaitë doesn’t support doubled curls. I would
favor Annatar for this reason, but as long as you note which font the
attested example requires, we can make it work. I am sure Halfdan would be
most pleased if Annatar worked most reliably. My preference in general is
to transcribe Elvish however, and very few elves, I’m sure, wrote with
Sauron’s own cursive.
I *did* have to modify Annatar for the transcriber. There are a small
number of symbols that were spread between Annatar and a supplementary
font. Web font rendering doesn’t allow one font to draw a diacritic over
another font, and the transcriber doesn’t require all of the glyphs to be
easily typed on an English keyboard layout, so I took some liberties.
…On Sun, Oct 13, 2019 at 12:29 PM dreamingfifi ***@***.***> wrote:
For general use English - there are actually remarkably few texts. Tolkien
preferred the "Northern Variety" for writing English, which is a full mode.
There is so little of General Use English that we use some of the ideas
from the full mode to fill in gaps in our knowledge. But, I can gather them
up as images, then write them out with a Dan Smith font (should I use
Parmaite or Annatar?)
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#24?email_source=notifications&email_token=AAAOXBW6P5M4Q7YXA2JE4XDQONZHXA5CNFSM4I65UNQ2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBC56EY#issuecomment-541450003>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAAOXBXLX24QROB6K7JX4OTQONZHXANCNFSM4I65UNQQ>
.
|
|
Okay, I will be using your notation then. I just spent an hour scanning images and hunting down what I could find of English General Use. I will cut these scans up and enhance the text if I can, then write it out with your notation. What's the file where you have a few broken down? Or, since I'm a novice at using your notation, would you rather I just post them here and you can work them into the page properly? |
|
The tests are in test/general-use.js, under english. If you would prefer to
post here, that’s still helpful.
…On Sun, Oct 13, 2019 at 1:21 PM dreamingfifi ***@***.***> wrote:
Okay, I will be using your notation then. I just spent an hour scanning
images and hunting down what I could find of English General Use. I will
cut these scans up and enhance the text if I can, then write it out with
your notation. What's the file where you have a few broken down? Or, since
I'm a novice at using your notation, would you rather I just post them here
and you can work them into the page properly?
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#24?email_source=notifications&email_token=AAAOXBSFHA4SLZP4CSYZPXTQON7K3A5CNFSM4I65UNQ2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBC7JHI#issuecomment-541455517>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAAOXBUYRXUGXBA4F6VWWB3QON7K3ANCNFSM4I65UNQQ>
.
|
|
Here's the first one:
Literally written like this:
Reproduced with Dan Smith font: Character-by-character description |
|
This is good material. Thank you. |
|
Literally written like this:
Note: O and U are switched, and there are a few spelling errors, since this is a rough draft. The lambe in the last word is drawn so long that it looks like it’s a closed lúva on the following tengwa, but by comparing it to the completed version, we can see the difference more easily. Reproduced with Dan Smith font: Character-by-character description:
Literally written like this:
Note: O and U are switched. Reproduced with Dan Smith font: Character-by-character description:
|
|
Literally written like this:
Note: I and E are switched in this text. Reproduced with Dan Smith font: (double O is written with “č” - I’m not sure if that is different in your font or not) Character-by-character description: |
|
It says:
Literally written like this:
Reproduced with Dan Smith font: Character-by-character description: |
|
Literally written like this: O and U are switched, and the vowel Y is used for the Y of “you.”
Reproduced with Dan Smith font: Character-by-character description: And that's everything. Do you need page numbers and references for these? |
|
By way of Reddit, found a sample of Christopher Tolkien’s variant mode used in the Silmarillion. Can also scan some of the interior title page passages from my personal copies and describe them.
https://www.etsy.com/listing/203926652/leather-covered-copy-of-the-silmarillion |
|
Sure! The reason I didn't before are because they aren't from Tolkien himself, and I consider them a secondary resource rather than a primary resource. |
|
I'd rather prioritize texts by J. R. R. Tolkien - of which there are a LOT. I've done all of the General use orthographic English, but I could do the "Northern Variety" orthographic English texts next, which is what the book after the next one I'm publishing will be about. |
|
That’s a good plan. My intent was to leave a note to self so I could close
a tab. You’ve already contributed enough material to drive another round of
development. Probably more than one! I have not finished copying these into
a working test harness, much less modifications necessary to pass those
tests.
…On Sun, Nov 24, 2019 at 12:04 PM Fiona Jallings ***@***.***> wrote:
I'd rather prioritize texts by J. R. R. Tolkien - of which there are a
LOT. I've done all of the General use orthographic English, but I could do
the "Northern Variety" orthographic English texts next, which is what the
book after the next one I'm publishing will be about.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#24?email_source=notifications&email_token=AAAOXBTCHZNPQ5DHPFGMUDLQVLM3HA5CNFSM4I65UNQ2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEFATTEA#issuecomment-557922704>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAAOXBVCFCW7S5U6YA65BGDQVLM3HANCNFSM4I65UNQQ>
.
|
|
Alright, would you like me to start? I can focus on the bigger, more famous texts and then go on to the more obscure ones. There are a lot more of the Northern Variety ones than there are of the others. |
There are a number of things we can do with pictures of attested samples.
For one, we can translate these to test cases for the appropriate mode. We can also create mode options to match these samples as closely as possible. Every scrap of Tolkien’s work has the potential to be a different incarnation of his ideas worthy of imitation.
Each of these also weighs different default behaviors differently, and we can take those into account specifically and generally. The mode for general use should be quite internally consistent, tuned to respect the most representative behaviors.
So it would be neat to collect scans of Tolkien’s example tengwar and capture them in the
publicdirectory, preferably with a name that matches their TE/VT-style reference name.The text was updated successfully, but these errors were encountered: