Re: Ipod touch owner on the fence
 

kinetic scrolling is an absolute need for a device with a touchscreen the size of NIT/ipod/iphone.
This I tell you, when the first time I've used kinetic scrolling is in microb webkit addon and tear.
The problem you have (and so do I), is an implementation "bug" in tear (remember myself to file a bug), link action should only happen after confirming there wasn't a drag action.

About the dpad, it is unusable in the n810 (damned designers playing games with keyboard alignment)

Re: Ipod touch owner on the fence
 

ah yes, i keep forgetting about them moving that dpad. another reason to hang on to my N800 ;)

Re: Ipod touch owner on the fence
 

I perceive kinetic scrolling as a sort of page-up/page-down. Some people don't use PgUp/PgDn, others do it all the time. Same for kinetic scrolling. From a usability standpoint I would prefer drag-zooming (<shameless plug>like in pyqtoreader</shameless plug>), but hey, kinetic looks cool, and if it's cool it must be good.

Re: Ipod touch owner on the fence
 

ah, yes, the rule of cool. that somehow explains the whole concept of apples marketing...

Re: Ipod touch owner on the fence
 

in fact: that's what a scrollbar is for. no matter how long a page is, you'll always get to any part of it with one single gesture in no time. no holding down the d-pad forever, no kinetic scrolling... again... again... still not there....ooops! too far.

it's efficient. that's even better than "cool". ;)

Re: Ipod touch owner on the fence
 

It is not cool, it is ergonomic. Kinetic makes very easy and efficient to move through a page bigger than the screen, by using a very natural hand/wrist movement.
Methods like pushing buttons on those tiny devices force the fingers to adopt non natural positions or angles while applying force, which in turn causes tendon strain, and pain in the end.
Non kinetic scrolling with the stylus like microb is very suboptimal, since moving through a big page takes a very long number of strokes (and microb scrolling responsivity is far from good). It also implies some tendon stress, since those prolonged strokes are done with more strength than quick strokes, and more strength is required to grip the device.
Finally, kinetic scrolling allows more feedback because the user sees a continuous flow of the page, with a speed response similar to what is expected in the "physical" world (that is why it seems "cool")

Re: Ipod touch owner on the fence
 

Not a huge fan of Kinetic scrolling. Half the time (from using it in Tear, Canola, Mediabox, or my cousin's iPhone) I have trouble getting it to go where I want. It's useful for long movements, but for short movements I prefer a scrollbar. But the ones in Maemo are to small so I used a finger friendly theme which makes scrollbars bigger.

Re: Ipod touch owner on the fence
 

It is cool. There is no need for the image to 'coast'. You could land at the extrapolated end position immediately, just as page up or page down does not need to animate the scroll. Kinetic, as in animated method of scrolling, is meant to be cool.

This is plain wrong. You cannot compare two specific implementations and draw conclusions about the physical interfaces in general. If it has a button, it must cause strain and if it's a stylus/finger it never does that ? Doesn't sound plausible to me at all.

Again, there are not just two ways of scrolling content. See pyqtoreader - it's faster and more precise than kinetic scrolling. Kinetic is just hero of the day, there are plenty other methods of scrolling (stylus based or other) that can replace it one day (given apple does a fancy implementation of them ;) )

Umm, and you cannot do continuous 'flow' scrolling with scrollbars, buttons, drag-zooms, or other methods ? I'm afraid I don't get it.

Re: Ipod touch owner on the fence
 

That's nonsense. Without actually seeing what happens, it would be extremely hard to learn the amount of force required to land at the desired position. Moreover, regularly you have no idea where you want to scroll to, so you want to see the contents while they pass by. Last but not least, the animation allows you to stop the scrolling, when you see that you arrived at the desired position.

Like most animations in modern GUIs, the "wow factor" is merely a byproduct. Useless animations don't impress anybody anymore. But some people regard every animation as useless, and that's just plain wrong. Fortunately, not everybody is a UI designer. :)


Be that as it may, that doesn't change the fact that panning in the Diablo browser is not as ergonomic as kinetic panning. In a sense, kinetic panning is nothing but a refinement to normal page panning, as it gives you more options while still allowing you to "drag pan" only.

I can't install pyqtoreader right now, so that is not telling me anything. It would certainly be interesting to discuss alternative methods of scrolling, and especially which of them may be as universally useful and intuitive as kinetic panning.

Re: Ipod touch owner on the fence
 

No way. Kinetic is meant to be usable. With page up/down you move the page by big integer units, so the extrapolated end position may easily end too up or too down, then you'll have to scroll more (key up/key down) at constant speed to center the content. It is very easy not to get to the desired position if you don't know where it is, too, leading to repeated searches up and down.

You are wrong, but may be I haven't expressed myself clearly enough.
By buttons here, I mean physical buttons, as n810 keys/dpad/etc, where the size, position and stiffness of those buttons is of great importance. What is more, we are talking here in the context of some particular devices, so I'm not going all general. The case of the n810 is a clear example, trying to use its keyboard, given the way the device must be hold and the fingers posture over the keyboard, will cause trouble after a relatively short time.
The use of GUI "buttons" is a different matter, since the strength needed in a touchscreen is lower. Here prolonged postural issues may be more relevant, depending on how the device is/must be hold and if/how the stylus is/must be used, influenced by usability issues like GUI element size, device/GUI responsiveness, repetitivity, etc.
May be there aren't just two ways, but of those I actually could test in the NIT/ipod/iphone, kinetic is the better. That is not saying it is just alone the best, since for example in very large pages it is more efficient to use other GUI method for coarse positioning, for example moving a box over a zoom-out view, and then kinetic scrolling/drag for fine positioning.
So, if a more usable/ergonomic method replaces it, I'll welcome it.

Have I said that? I'm afraid not. What I said is a quite different thing, read the sentence as a whole.
About scrollbars, on a device like NITs/ipod/iphone it is imprecise and jerky (stylus positioning, long pages), takes screen real state, and element size and/or position may make them really difficult to use. GUI buttons have some of those issues too. Hard buttons, see issues about tendon strain.
About "drag-zooms" (drawing a box by clicking and then dragging the opposite vertex and zooming the box content), I fail to see how it relates to scrolling.

About your pyqtoreader, unfortunately I'm not able to install it (some issue with python-qt4-* package versions that apt-get doesn't like), so I cannot comment on the scrolling methods you have implemented.