On Thu, 27 Apr 1995 13:00:10 CDT Patrice Bonhomme said:
>But today i need to encode strip cartoons into SGML using the TEI-P3.
>Is Someone try to do this before.
>I want to parse the comic strips in pages, pictures and dialogue.
>
>For example :
> ...
No one has used TEI to encode comic strips that I know of. You can
certainly modify the DTD to include the PICTURE and PAGE elements
you suggest you want, but you can also do much the same thing with
DIV elements (call them TYPE=PICTURE or TYPE=PANEL) and PB (page break).
Using these elements, and using Q for the dialog, you might have
something like this; as you see, it's very similar to the skeleton
you posted, but requires no extensions to the published DTD.
<!-- all EMPH phrases are slightly bold in the source. -->
<!-- all NAME elements are in double quotes. Alternate tagging as
SOCALLED also plausible -->
<!-- all dialog is all caps in the strip; case distinctions have
been introduced by the transcriber for ease of reading. -->
<body>
<pb n=1 id=page1>
<div type='strip' id=s1>
<div type='panel' id=p1>
<q id=p1s1 who='Griffy'>Mr. Toad, do you worry about your
<emph>waistline?</emph></q>
<q id=p1s2 who='Toad' >The toad does not <emph>worry</>.
The toad <emph>ruminates</emph>.</q>
</div>
<div id=p2>
<q id=p2s1 who='Griffy'>I thought <emph>tranquility</>
would come with <emph>age</> .... but
I'm as <emph>anxious</emph> as
ever ...</q>
<q id=p2s2 who='Toad' >I no longer remove th' <emph>pickle</>
from my <name>Whopper</>.</q>
</div>
<div id=p3>
<q id=p3s1 who='Griffy'>Really? Does that mean you
<emph>reject</> th' entire tradition
of western <emph>individualism?</emph></q>
<q id=p3s2 who='Toad' ><emph>Control</> is an illusion.
<emph>Embrace</> th' <emph>pickle.</emph>
</q>
</div>
</div>
<pb n=2 id=page2>
<div type='strip' id=s2>
<div type='panel' id=p4>
<q id=p4s1 who='Griffy'>What's it <emph>doing</> outside, Zippy?
Anything <emph>coming down</> yet?</q>
<q id=p4s2 who='Zippy' >Tiny toy tubas.</q>
</div>
<div id=p5>
<q id=p5s1 who='Zippy' ><emph>Tiny toy tubas</> are coming down.
<emph>Thousands & thousands</> of
tiny toy tubas.</q>
<q id=p5s2 who='Zippy' >Coming down.</q>
<!-- two balloons for Zippy in same panel. -->
</div>
<div id=p6>
<q id=p6s1 who='Griffy'>What is th' <emph>sound</emph> of
thousands of <emph>tiny toy tubas</>
coming down?</q>
<q id=p6s2 who='Zippy' ><emph>Oom pah-pah,</> tinkle.
<emph>Oom pah-pah,</> tinkle.
<emph>Oom pah-pah,</> tinkle.
</q>
</div>
</div>
<!-- ... -->
</body>
(Note: these examples are from the comic strip Zippy, by Bill Griffith,
which the Washington Post recently described as "the comic strip that
fulfills American newspaper readers' minimum daily surrealism
requirements." They are copyright 1995 by Bill Griffith, and I am
counting on him either not to see this or not to mind. As far as I can
tell, these are the strips for January 7 and 10, 1995.)
The identifiers on the DIV and Q elements would allow links between the
transcriptions and scanned images of the strip itself. If it is important
to record the disposition of panels on the page, I would use the REND
attribute on each panel to give dimensions and location on the page.
The LINKGRP and LINK elements could also be used, in conjunction with XPTR
elements pointing into the image, to accomplish the same thing. An
example is given in section 14.4.3 of the Guidelines. Which method to
use depends on how elaborate you want your encoding of the layout to be.
Complex layout of the type found in some French and Japanese (less often
in American) comic books / BDs / manga might well require very complex
markup to capture in detail.
On the other hand, if you are just interested in capturing the text,
e.g. for studies of colloquial language, you can get by with much less.
The major drawbacks I can see to transcriptions of this sort are (1) DIV
is less intuitively obvious than STRIP or PANEL, and (2) the information
on the disposition of the panels is not standardized, as the Guidelines
do not prescribe methods of recording this information. I hope that we
will be able to fund further technical work to provide a basic vocabulary
for describing the disposition of elements on the page; it remains an
open problem.
Are there other suggestions for encoding comic strips?
-C. M. Sperberg-McQueen
ACH / ACL / ALLC Text Encoding Initiative
University of Illinois at Chicago
[log in to unmask] / u35395@uicvm
|
