Re: [9fans] panning or scrolling, page(1)

Re: [9fans] panning or scrolling, page(1)

[rog]

[-- Attachment #1: Type: text/plain, Size: 1389 bytes --]

i like the panning model too, but i don't tend to view very large
images.

when you've got a large image (many times the size of the area you can
see), panning can be a bit of a problem because you can only go a small
distance at once.

i.e. if i'm at the bottom left of a large image and i want to go to the top
right, it might take me 10 or so "click & pan" mouse movements to get
there.

but for images that size, maybe a different interface is called for
anyway.

my last job was writing radio propagation prediction software, which
involved the viewing of large maps (e.g. the entire of the UK with one
height sample every 100m). i once accidentally created quite a nice
viewing interface when i set the map to be always centred around the
transmitter.

normally, you could drag transmitters around with the mouse, and i
hadn't yet disabled this, so when i dragged the transmitter, it moved,
but the map kept up, giving the nice sensation of flying over the
landscape. (as long as i kept wiggling the mouse...)

maybe that kind of interface might sit better with the panning that
page uses (e.g. middle button "marks the spot", and map keeps panning
at a speed proportional to the vector from the spot to the current
mouse position).  i always find it a little disconcerting that the
middle button quits page so abruptly anyway.

  cheers,
    rog.


[-- Attachment #2: Type: message/rfc822, Size: 1671 bytes --]

To: cse.psu.edu!9fans
Subject: Re: [9fans] panning or scrolling, page(1)
Date: Fri, 2 Mar 2001 21:49:26 -0500
Message-ID: <20010303024929.04C5E199DC@mail.cse.psu.edu>

I strongly prefer the panning model to the scrolling model. It's so much
easier to use I wish it were available everywhere.  Are you really asking for
some indication of what part of the image is visible?  Scroll bars provide
that at the cost of a crappy interface, but there's no reason at all why a
panning interface can't give you some indication of what subset of the
entire object you're viewing.

-rob

Re: [9fans] panning or scrolling, page(1)

[okamoto]

>I strongly prefer the panning model to the scrolling model. 

Yes, I had thought you'd done it already, which made me hesitate to open
this discussion.  ^_^

>  Are you really asking for
>some indication of what part of the image is visible?  

yes, I do, for images we are dealing with.  Let's think of a Mars Global Surveyor 
image, which usually has a very vertically long rectangular size, say such that
1024x5632.   In such case, we can never see it in a window, but only in some
part a once.

If we use panning model without any indication of position information (page91), 
it's not easy to image what part we are now seeing, which leads us some difficulty
to get that whole image in our mind.  Viewing a part always with getting its whole 
image is very important to us to find something from that image.   Therefore, we 
added vertical/horizontal scrollbars here, which enables us to get the idea 
1)what size of the whole image, 2) what position we are now seeing.

>that at the cost of a crappy interface, but there's no reason at all why a
>panning interface can't give you some indication of what subset of the

We've also considered the possiblity, with panning interface,  to have some 
small woindow which shows us the whole image size and its position.   
However, I don't see any reason this is a better idea than scrollbars...

In the case of documents, those have position information in themselves, and
we don't need such in our user interface.   This is the reason why I thought
page(1) was originally designed for documents.

Kenji

Re: [9fans] panning or scrolling, page(1)

[Fariborz 'Skip' Tavakkolian]

Like the mouse cursor when the panning button is pressed? It would be where
the user is looking. 

>some indication of what part of the image is visible?  Scroll bars provide
>that at the cost of a crappy interface, but there's no reason at all why a
>panning interface can't give you some indication of what subset of the
>entire object you're viewing.

Re: [9fans] panning or scrolling, page(1)

[rob pike]

I strongly prefer the panning model to the scrolling model. It's so much
easier to use I wish it were available everywhere.  Are you really asking for
some indication of what part of the image is visible?  Scroll bars provide
that at the cost of a crappy interface, but there's no reason at all why a
panning interface can't give you some indication of what subset of the
entire object you're viewing.

-rob

[9fans] panning or scrolling, page(1)

[okamoto]

I've been hesitated to post this here...  I decided now to post here, because
someone else may have differnt opinion on this issue.

Page(1) applies panning to see a part of images, but not scrolling.  I've been
thought this is because page(1) was originally designed to read documents
formatted in postscript or pds, and expanded to many kinds of images.

In our work, we are mainly using page to see satellite images of Mars etc..
In this case, we can never predict what part of the whole image we are now
seeing from the image itself.  For such circumstance, scrolling image horizontally
or vertically is more profittable user interface.  We have now page2 program 
locally for this purpose, which is coded mainly by Yoshitatsu.  (We can now
view, on Plan 9, most of NASA's satellite images, such as Viking, Voyager, 
Magellan, Clementine, Mars Pathfinder and Mars Global Surveyor. ^_^).   
However, when we read documents by page, panning method is fancier to 
me, too.

How about your opinion?

Kenji