|
| Authored by: Gringo_ on Wednesday, October 24 2012 @ 08:35 AM EDT |
| ! [ Reply to This | Parent | # ]
|
|
|
| Authored by: Anonymous on Wednesday, October 24 2012 @ 08:37 AM EDT |
Aside from his terminology problem, three classes of
solution to this problem are apparent.
1) Provide a navigational overlay on top of the surface
being scrolled and scaled. For example a dynamic grid
overlay (at a scale set to ensure that at least one line
shows in X and Y axis on the visible portion). The scale and
position could be indicated with numbers alongside the grid,
or by different coloured lines.
Overlaying lines on top of content may be unacceptable in
many cases.
2) Preventing the user from scrolling past the end of the
valid data.
This is mainly appropriate in bounded non-sparse data, and
would be potentially unsuitable for geographic data. This
also leads to the dead screen UI problem, where the screen
doesn't scroll when the user is interacting with it.
3) Ensuring that if the user does scroll out of the valid
data area, the UI returns them to it. (the route the patent
took)
Three basic obvious classes of solution. You either prevent
the user from leaving the area (2), ensure that when they
leave they are sent back to acceptable positions (3), or you
make the "desert fog" navigable so that they can find their
own way back (1).
All of them are so obvious as to make a patent seem
ludicrous (if not then I'm starting a troll company), so it
just comes down to the implementation details. If I need a
patent to teach me any of these approaches then I also
probably need a lobotomy.
Disregarding the prior art, surely returning the content
scroll to an acceptable value using an elastic metric is
also simple and completely obvious. The first obvious way
would be to interpolate between the current value and the
acceptable value linearly, but would swiftly become apparent
that an elastic approach would be superior due to giving
better feedback about the correct end position, and allowing
rapid correction of big overshoot values whilst not being
too fast to see the animation for short values.[ Reply to This | Parent | # ]
|
|
|
| Authored by: jesse on Wednesday, October 24 2012 @ 08:56 AM EDT |
Back around 1987 the solution was to bring up a list of open windows... Even if
the window was off the screen, it would still be listed.
When that entry was selected the window would be repositioned and in view.
So, yes, it had been solved.[ Reply to This | Parent | # ]
|
|
|
| Authored by: StormReaver on Wednesday, October 24 2012 @ 09:28 AM EDT |
> ...but his characterization of this issue as "desert fog" is a
bit simplistic and his assertion that no-one fixed this is a little bit
disingenuous.
That is being kind. The reason that no one had ever filed a patent on solving
the problem is because the problem is so trivial to solve, any first-semester
programming student would have readily arrived at the same (or strongly similar)
solution.
Putting it bluntly, the problem is trivially simple. But then, that seems to be
the M.O. of technology companies nowadays: patent every trivial operation that
has been in existence for 20 years, and market it as an astounding
breakthrough.
[ Reply to This | Parent | # ]
|
|
|
| Authored by: Anonymous on Wednesday, October 24 2012 @ 12:49 PM EDT |
In the late 80s Amiga users used to make fun of Windows users
for this.
Workbench (the graphical shell on amiga) did not allow
windows to go off-screen so you could never lose the window.
This clearly made amiga superior and this was repeatedly
pointed out to anyone that used a different computer.
[ Reply to This | Parent | # ]
|
|
