217 |
|
|
218 |
Buoys are a user interface technique invented within the Fenfire |
Buoys are a user interface technique invented within the Fenfire |
219 |
project. |
project. |
220 |
|
What we call *buoy* is a commonly used tool in technical diagrams: |
221 |
The placement of buoys is governed by simple geometric rules. |
placing a label at the edge of the image and connecting the label |
222 |
|
to the relevant location (anchor) by a line (see Fig. [ref-fignasa]_). |
223 |
*Nadir* |
|
224 |
|
The placement of buoys is governed by simple geometric rules |
225 |
|
designed to show the most relevant related information close |
226 |
|
to the focus of the view. |
227 |
|
|
228 |
|
.. *Nadir* rotations is a related technique which can be used to |
229 |
|
make the buoys visually distinct by rotating them |
230 |
|
so that their virtual y-axis points toward a *nadir* |
231 |
|
at approximately one screen height below the physical screen. |
232 |
|
|
233 |
The research goals related to buoys are first of all to characterize |
The research goals related to buoys are first of all to characterize |
234 |
the qualitative and quantitative differences between |
the qualitative and quantitative differences between |
264 |
the structure. |
the structure. |
265 |
|
|
266 |
We use buoys as link targets floating around the focus. |
We use buoys as link targets floating around the focus. |
|
What we call *buoy* is a commonly used tool in technical diagrams: |
|
|
placing a label at the edge of the image and connecting the label |
|
|
to the relevant location (anchor) by a line (see Fig. [ref-fignasa]_). |
|
267 |
We research the geometry and layout of the buoy placement |
We research the geometry and layout of the buoy placement |
268 |
for coherent views and animation. |
for coherent views and animation. |
269 |
|
|
280 |
(wiggly freehand line, see Fig. [ref-fignasa]_). |
(wiggly freehand line, see Fig. [ref-fignasa]_). |
281 |
We apply this technique by drawing the buoys as |
We apply this technique by drawing the buoys as |
282 |
non-photorealistical pieces torn off the target document. |
non-photorealistical pieces torn off the target document. |
283 |
|
|
284 |
To allow for fluid animation, |
To allow for fluid animation, |
285 |
the shapes of the break lines need to be carefully designed. |
the shapes of the break lines need to be carefully designed. |
286 |
|
For example, animating a fragment to a full document |
287 |
|
should not look like the edge just gliding |
288 |
|
over the document, but rather as if larger and |
289 |
|
larger parts were magically torn off the original document. |
290 |
|
|
291 |
.. figure:: ../../manuscripts/xupdf/mercury5part2 |
.. figure:: ../../manuscripts/xupdf/mercury5part2 |
292 |
:width: 8.45cm |
:width: 8.45cm |
298 |
freehand lines are drawn to indicate that |
freehand lines are drawn to indicate that |
299 |
the depicted object extends beyond the section shown. |
the depicted object extends beyond the section shown. |
300 |
|
|
301 |
The concrete research goals are to perform usability experiments |
The concrete research goals are to evaluate the practical value of |
302 |
to evaluate the practical value of break lines as implemented |
break lines as implemented in the Fenfire project and |
303 |
in the Fenfire project. Especially the properties of the animation ... |
to study more formally the perceptual properties |
304 |
|
of different types of torn edges. |
305 |
|
Especially the properties of the animation are interesting, |
306 |
|
because it looks somewhat natural even though |
307 |
|
there is no real-world analogue. |
308 |
|
|
309 |
|
|
310 |
Fillets |
Fillets |
311 |
''''''' |
''''''' |
344 |
JVK, HH |
JVK, HH |
345 |
|
|
346 |
Most user interfaces only use pre-rendered character |
Most user interfaces only use pre-rendered character |
347 |
bitmaps for text, which forces text to be |
bitmaps of the font, which forces text to be |
348 |
rendered on straight horizontal lines with constant size |
drawn on straight horizontal lines with constant size |
349 |
scaled in discrete steps. |
scaled in discrete steps. |
350 |
|
|
351 |
The Fenfire interface uses text more freely than conventional |
The Fenfire interface uses text more freely than conventional |
352 |
user interfaces, applying rotation, and fisheye transforms |
user interfaces, applying rotation, fisheye distortion and |
353 |
|
smooth animation. |
354 |
|
|
355 |
Rendering text using the 3D hardware is problematic because |
Rendering text using the 3D hardware is problematic because |
356 |
the texturing algorithms are mainly designed for full-color image |
the texturing algorithms are mainly designed for full-color image |
362 |
of per-pixel operations has been increased considerably; it is now |
of per-pixel operations has been increased considerably; it is now |
363 |
possible to run small *fragment programs* for each rendered pixel. |
possible to run small *fragment programs* for each rendered pixel. |
364 |
This flexibility allows us to overcome the restrictions in the |
This flexibility allows us to overcome the restrictions in the |
365 |
|
texture filtering. |
366 |
|
|
367 |
However, designing such filters in an *ad hoc* fashion is |
However, designing such filters in an *ad hoc* fashion is |
368 |
extremely cumbersome and error-prone. What is needed is a suitable |
extremely cumbersome and error-prone. What is needed is a suitable |
369 |
mathematical framework for |
mathematical framework for modeling the |
370 |
|
*perceptual* qualities of rendered text. |
371 |
|
|
372 |
The goals in this area are the development of the mathematical framework |
The goals in this area are the development of the mathematical framework |
373 |
for qualitatively |
for qualitatively |