[linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
24 Aug
2004
24 Aug
'04
5:26 p.m.
I just realized I posted this to LAD but not LAU. Since the whole point
of HIGs is to be reflect user rather than developer needs, I am
reposting it here.
OK, so if I were to start writing the Linux Audio Human Interface
Guidelines based on all of the feedback from this post, I would say
something like the following. I have used the type of language you
would find in an Internet RFC.
"Sliders MUST be operable with the mouse wheel. This MUST only be
enabled when mousing over a slider, otherwise, the mouse wheel MUST
retain its normal behavior of scrolling the window.
For a vertical slider, rolling the mouse button away from you
(hereafter: wheel-up event) MUST increase the value of the parameter,
and wheel-down MUST decrease the value.
For a horizontal slider, wheel-up SHOULD increase the value, and
wheel-down SHOULD decrease the value by default. However, this behavior
MUST be configurable. If a system-wide facility to control
this is unavailable then the app MUST provide a configuration
mechanism."
Additional submissions are welcome and in fact essential, since the
above only covers one aspect of one widget's behavior.
Lee
25 Aug
25 Aug
11:27 a.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
Last Tuesday 24 August 2004 20:10, Lee Revell was like:
I just realized I posted this to LAD but not LAU.
Since the whole point
of HIGs is to be reflect user rather than developer needs, I am
reposting it here.
OK, so if I were to start writing the Linux Audio Human Interface
Guidelines based on all of the feedback from this post, I would say
something like the following. I have used the type of language you
would find in an Internet RFC.
"Sliders MUST be operable with the mouse wheel. This MUST only be
enabled when mousing over a slider, otherwise, the mouse wheel MUST
retain its normal behavior of scrolling the window.
For a vertical slider, rolling the mouse button away from you
(hereafter: wheel-up event) MUST increase the value of the parameter,
and wheel-down MUST decrease the value.
For a horizontal slider, wheel-up SHOULD increase the value, and
wheel-down SHOULD decrease the value by default. However, this behavior
MUST be configurable. If a system-wide facility to control
this is unavailable then the app MUST provide a configuration
mechanism."
YES.
Additional submissions are welcome and in fact
essential, since the
above only covers one aspect of one widget's behavior.
Sure, but if other widget behaviours could be agreed upon, defined like this
and implemented universally, I will be one happy user. :-)
cheers
tim hall
12:42 p.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
On Wed, Aug 25, 2004 at 02:11:45PM +0100, tim hall wrote:
Last Tuesday 24 August 2004 20:10, Lee Revell was
like:
> Additional submissions are welcome and in fact essential, since the
> above only covers one aspect of one widget's behavior.
Here's one that I suppose is non-controversial:
Meters and Data Plots
Vertical meters should be oriented such that upward movement represents
increase in the metered value and down represents decrease.
Horizontal meters should be oriented such that rightward movement
represents increase in the metered value and left represents decrease.
Analog VU-style meters are considered primarily horizontal.
3D meters and data plots should be oriented like a cartesian plane:
up and right represents increase, down and left represents decrease.
--
Paul Winkler
http://www.slinkp.com
1:11 p.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
On Wed, Aug 25, 2004 at 10:26:55 -0400, Paul Winkler wrote:
On Wed, Aug 25, 2004 at 02:11:45PM +0100, tim hall
wrote:
Heh, non-controversial you say ;)
One convention is that reduction meters got right to left (and presumably
top to bottom, though I dont think I've never seen a vertical one).
This is neccesary if the reduction measurement (attenuation etc.) is to
have the same scale as a normal measurment. c.f. JAMin's compressor gain
reduction meters: http://www.ecs.soton.ac.uk/~swh/jamin-log-spectrum.png
somewhat old screenshot, gut you can see the right-to-left meters and the
bottom of each compressor section.
- Steve
Last Tuesday 24 August 2004 20:10, Lee Revell was
like:
> Additional submissions are welcome and in fact essential, since the
> above only covers one aspect of one widget's behavior.
Here's one that I suppose is non-controversial:
Meters and Data Plots
Vertical meters should be oriented such that upward movement represents
increase in the metered value and down represents decrease. 
2:50 p.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
On Wed, Aug 25, 2004 at 03:55:47PM +0100, Steve Harris wrote:
On Wed, Aug 25, 2004 at 10:26:55 -0400, Paul Winkler
wrote:
GAH... I totally forgot about that. I just looked at my RNC and
you're right :-P It has an LED meter like so:
+----- gain reduction ------+
20 16 12 8 6 4 2 0
* * * * * * * *
I could swear I remember seeing an old tube compressor that
had a gain reduction meter with "more" to the right, so it
moved in opposition to the corresponding VU meter.
No idea what model it was, and this was about 6 years ago
so I could be mistaken.
I wonder how consistent this is in the world of hardware?
maybe I'll pop a question on rec.audio.pro.
--
Paul Winkler
http://www.slinkp.com
On Wed, Aug 25, 2004 at 02:11:45PM +0100, tim
hall wrote:
Heh, non-controversial you say ;)
One convention is that reduction meters got right to left (and presumably
top to bottom, though I dont think I've never seen a vertical one). Last Tuesday 24 August 2004 20:10, Lee Revell was
like:
> Additional submissions are welcome and in fact essential, since the
> above only covers one aspect of one widget's behavior.
Here's one that I suppose is non-controversial:
Meters and Data Plots
Vertical meters should be oriented such that upward movement represents
increase in the metered value and down represents decrease. 
3:01 p.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
Paul Winkler wrote:
+----- gain reduction ------+
20 16 12 8 6 4 2 0
* * * * * * * *
I may have this backwards, but, iiuc, higher output is still on the
right. If so, it would seem to still meet the standard if expressed as
something like "higher levels of output to the right".
(Higher numbers of gain reduction on the left correspond to lower levels
of output -- not??)
Randy Kramer
3:48 p.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
On Tue, Aug 24, 2004 at 10:54:55PM -0400, Ruth A. Kramer wrote:
Paul Winkler wrote:
that might do.
+----- gain reduction ------+
20 16 12 8 6 4 2 0
* * * * * * * *
I may have this backwards, but, iiuc, higher output is still on the
right. If so, it would seem to still meet the standard if expressed as
something like "higher levels of output to the right". (Higher numbers of gain reduction on the left
correspond to lower levels
of output -- not??)
yep.
--
Paul Winkler
http://www.slinkp.com
4:12 p.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
Correcting myself:
On Wed, Aug 25, 2004 at 01:32:55PM -0400, Paul Winkler wrote:
On Tue, Aug 24, 2004 at 10:54:55PM -0400, Ruth A.
Kramer wrote:
No no no! See how confusing this all is? :-(
As the input signal increases, the gain is reduced more.
So the meter moves to the left.
With silent input, there is no gain reduction, so the meter will be all
lit up (i.e. "to the right").
--
Paul Winkler
http://www.slinkp.com
Paul Winkler wrote:
that might do.
+----- gain reduction ------+
20 16 12 8 6 4 2 0
* * * * * * * *
I may have this backwards, but, iiuc, higher output is still on the
right. If so, it would seem to still meet the standard if expressed as
something like "higher levels of output to the right". (Higher numbers of gain reduction on the left
correspond to lower levels
of output -- not??)
yep.
4:33 p.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
Paul Winkler wrote:
No no no! See how confusing this all is? :-(
As the input signal increases, the gain is reduced more.
So the meter moves to the left.
With silent input, there is no gain reduction, so the meter will be all
lit up (i.e. "to the right").
Paul,
Well, I guess I see how confusing it is. ;-) Having never fooled with a
gain reduction device (alafaik), it may take some time to get my head
around this. But, it will definitely be a back burner effort.
regards,
Randy Kramer
5:06 p.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
On Wednesday 25 August 2004 01:56 pm, Paul Winkler wrote:
Correcting myself:
On Wed, Aug 25, 2004 at 01:32:55PM -0400, Paul Winkler wrote:
Passes a threshold. It's displaying the diff between what is and what should
never be.
On Tue, Aug 24, 2004 at 10:54:55PM -0400, Ruth A.
Kramer wrote:
No no no! See how confusing this all is? :-(
As the input signal increases, the gain is reduced more.
So the meter moves to the left.
Paul Winkler wrote:
that might do.
+----- gain reduction ------+
20 16 12 8 6 4 2 0
* * * * * * * *
I may have this backwards, but, iiuc, higher output is still on the
right. If so, it would seem to still meet the standard if expressed as
something like "higher levels of output to the right". (Higher numbers of gain reduction on the left
correspond to lower
levels of output -- not??)
yep. With silent input, there is no gain reduction, so the
meter will be all
lit up (i.e. "to the right").
With input up to the threshold. The ratio is what determines the amount of
gain reduction applied. You can pin it with 1dB if the ratio is high enough
and the threshold low enough.
It's showing you how much the gain reduction is pushing back, so being
in opposition to how strong you're pushing, it's not unintuitive.
Anyway, this is the precedent, so no point in thinking too much about it.
5:51 p.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
On Wed, Aug 25, 2004 at 01:56:11 -0400, Paul Winkler wrote:
No no no! See how confusing this all is? :-(
As the input signal increases, the gain is reduced more.
So the meter moves to the left.
With silent input, there is no gain reduction, so the meter will be all
lit up (i.e. "to the right").
Or all dark, depending on which way the pointer moves.
- Steve
(Higher numbers of gain reduction on the left
correspond to lower levels
of output -- not??)
yep.
4:55 p.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
On Tuesday 24 August 2004 10:54 pm, Ruth A. Kramer wrote:
Paul Winkler wrote:
The right side is still the ceiling, yes.
+----- gain reduction ------+
20 16 12 8 6 4 2 0
* * * * * * * *
I may have this backwards, but, iiuc, higher output is still on the
right. If so, it would seem to still meet the standard if expressed as
something like "higher levels of output to the right".
(Higher numbers of gain reduction on the left
correspond to lower levels
of output -- not??)
Randy Kramer
3:42 p.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
On Wed, Aug 25, 2004 at 12:34:19 -0400, Paul Winkler wrote:
GAH... I totally forgot about that. I just looked at my RNC and
you're right :-P It has an LED meter like so:
+----- gain reduction ------+
20 16 12 8 6 4 2 0
* * * * * * * *
I could swear I remember seeing an old tube compressor that
had a gain reduction meter with "more" to the right, so it
moved in opposition to the corresponding VU meter.
No idea what model it was, and this was about 6 years ago
so I could be mistaken.
Could well, be I said "one convention", poor attempt at sarcasm.
Here's one that I suppose is non-controversial:
Meters and Data Plots
Vertical meters should be oriented such that upward movement represents
increase in the metered value and down represents decrease.
Heh, non-controversial you say ;)
One convention is that reduction meters got right to left (and presumably
top to bottom, though I dont think I've never seen a vertical one). I wonder how consistent this is in the world of
hardware?
maybe I'll pop a question on rec.audio.pro.
I think that (most of) the meter specifications say how the numbers should
be layed out, not which way the meter indicator moves.
- Steve
4:52 p.m.
New subject: [linux-audio-user] Linux Audio Human Interface Guidelines 0.01 (was Re: mouse wheel behavior and RFC: human interface guidelines)
On Wednesday 25 August 2004 10:55 am, Steve Harris wrote:
On Wed, Aug 25, 2004 at 10:26:55 -0400, Paul Winkler
wrote:
Yup. I think there's room for that one.
On Wed, Aug 25, 2004 at 02:11:45PM +0100, tim
hall wrote:
Heh, non-controversial you say ;)
One convention is that reduction meters got right to left (and presumably
top to bottom, though I dont think I've never seen a vertical one).
Last Tuesday 24 August 2004 20:10, Lee Revell was
like:
> Additional submissions are welcome and in fact essential, since the
> above only covers one aspect of one widget's behavior.
Here's one that I suppose is non-controversial:
Meters and Data Plots
Vertical meters should be oriented such that upward movement
represents increase in the metered value and down represents decrease. This is neccesary if the reduction measurement
(attenuation etc.) is to
have the same scale as a normal measurment. c.f. JAMin's compressor gain
reduction meters: http://www.ecs.soton.ac.uk/~swh/jamin-log-spectrum.png
somewhat old screenshot, gut you can see the right-to-left meters and the
bottom of each compressor section.
- Steve
7448
days inactive
7449
days old
linux-audio-user@lists.linuxaudio.org
13 comments
6 participants
participants (6)
-
John Check -
Lee Revell -
Paul Winkler -
Ruth A. Kramer -
Steve Harris -
tim hall