[tei-council] [TEI-DIR-WG] Rotation and Reflection

[Martin Holmes]

Sorry, sent this to the wrong list. Especially ironic, given the content...

On 12-12-14 10:51 PM, Martin Holmes wrote:
> Hi Rob,
>
> This is a publicly-archived list, so if that login is sensitive, you
> might want to change it.
>
> Nice XML! It's good to see it available. Not enough sites do that.
>
> Cheers,
> Martin
>
> On 12-12-14 06:10 PM, Robert Whalen wrote:

>> http://digitaltemple.rotunda.upress.virginia.edu
>>
>> Go to parallel display and find Easter Wings using one of the drop
>> downs. The rest is pretty intuitive. Note the link to the TEI-XML.
>>
>> Cheers,
>> Rob
>>
>> On 12/14/12 12:12 PM, Martin Holmes wrote:
>>>
>>>
>>> On 12-12-14 02:17 AM, Rob Whalen wrote:
>>>> On Dec 13, 2012, at 4:29 PM, Martin Holmes <mholmes at uvic.ca> wrote:
>>>>
>>>>> I'd like to bring in Robert and Stella here. Robert, does it seem
>>>>> to you that what's happening in Easter Wings is something very
>>>>> different from what's happening when Japanese is written (as it
>>>>> normally is, in a lot of contexts) from top-to-bottom, left to
>>>>> right?
>>>>
>>>> Clearly what's happening in ewings is entirely different. From a
>>>> writing or reading perspective, it's not an instance of t to b and l
>>>> to r but rather a block of text rotated clockwise 90 degrees. The
>>>> question for me is how best to capture this in a TEI-XML
>>>> transcription so as NOT semantically to suggest a t to b reading
>>>> orientation, and to render the rotated block in the display. An added
>>>> wrinkle, as concerns the encoded transcription, is that of the three
>>>> witnesses to the poem (encoded using TEI parallel segmentation), only
>>>> one is rotated. I'd like to preserve the line by line parallel
>>>> encoding while capturing the rotated orientation of the one witness
>>>> and writing XSL and CSS to handle it effectively.
>>>
>>> This is an interesting wrinkle that I hadn't really thought about, but
>>> now you raise it, it's obvious: whatever we choose to do to represent
>>> both text directionality and rotational phenomena must be expressible
>>> in ways which make it easy to connect it with specific witnesses or
>>> readings. Your encoding should be one of our test cases here.
>>>
>>> Cheers,
>>> Martin
>>>
>>>>> Stella, are there any instances in which Arabic might be written
>>>>> vertically? I don't recall seeing anything like that, but with shop
>>>>> signs etc. I guess it might happen. If it does happen, what is the
>>>>> effect on the shapes of glyphs? IIRC the shape of a glyph in Arabic
>>>>> is dependent on context, and enables the clean joining of glyph
>>>>> sequences.
>>>>>
>>>>> It occurs to me that one of our test cases for encoding complex
>>>>> rotational phenomena might be a captcha image; these often show
>>>>> letters partially rotated, and sometimes skewed in ways which mimic
>>>>> 3D rotation.
>>>>>
>>>>> Which brings me to a more general point: I'd like to start
>>>>> gathering a set of small but interesting test cases -- real
>>>>> documents, as short as we can find them, which cover all the
>>>>> phenomena we need to deal with. Easter Wings is one. It would be
>>>>> great to get examples of all the Japanese variants, Greek written
>>>>> right-to-left (I didn't even know that happened until recently),
>>>>> and English written vertically with letters both upright and not.
>>>>> If you have examples of texts that would be useful, especially if
>>>>> you have a page-image, and _especially_ if there are no
>>>>> restrictions on use of same so we could actually consider putting
>>>>> it in the Guidelines, please send them along.
>>>>>
>>>>> Cheers, Martin
>>>>>
>>>>> On 12-12-13 12:31 PM, Marcus Bingenheimer wrote:
>>>>>> Hi Martin,
>>>>>>
>>>>>> I like the logic of replacing reflection with rotation. Apart
>>>>>> from Ockham's razor it opens up new games like: What is this: ".
>>>>>> ._ _ _  _ _ _"" (answer: Non-serif "I love you" rotated 90 deg on
>>>>>> the x-axis). We should default the value for x and y rotation to
>>>>>> 180 deg, however, as most writing happens in 2D and the
>>>>>> introduction of 3D opens a can of worms. We do not want people to
>>>>>> think that we want to describe character "thinkness" or the
>>>>>> spherical qualities of an "o" ring with TEI. To say nothing of 4D
>>>>>> animations, let's leave that to TEI P8.
>>>>>>
>>>>>> I also like the idea of distinguishing between "rotational
>>>>>> effects" and "directionality". One can indeed argue that some
>>>>>> directions are merely a secondary effect introduced by a
>>>>>> typesetter or an artist, and that writing system have a innate
>>>>>> direction (or two equally valid ones as in Ch or Jp). For
>>>>>> instance, the Mongolian word in the Unicode example (Sec.3.3) is
>>>>>> not "actually" written that way, I would say that it is written
>>>>>> top-to-bottom and then rotated counter-clockwise 90deg.
>>>>>>
>>>>>> all the best
>>>>>>
>>>>>> marcus
>>>>>>
>>>>>>
>>>>>> -- Dr. Marcus Bingenheimer 馬德偉 Department of Religion, Temple
>>>>>> University http://mbingenheimer.net
>>>>>
>>>>> -- Martin Holmes University of Victoria Humanities Computing and
>>>>> Media Centre (mholmes at uvic.ca)
>>>
>>
>> --
>>
>> Robert Whalen, PhD
>>
>> Professor, Department of English
>>
>> Northern Michigan University
>>
>> 1401 Presque Isle Avenue
>>
>> Marquette, MI 49855
>>
>> 906-227-2678
>>
>> Office hours: Monday and Wednesday, 8:10-9:50 and 4:40-5:30
>>
>> rwhalen at nmu.edu <mailto:rwhalen at nmu.edu>
>>
>> http://myweb.nmu.edu/~rwhalen/home.htm
>> <http://myweb.nmu.edu/%7Erwhalen/home.htm>
>>
>> /The truth will set you free. But not until it’s finished with you./
>>
>> - David Foster Wallace
>>
>

-- 
Martin Holmes
mholmes at uvic.ca
UVic Humanities Computing and Media Centre