MOTM

We have this discussion from time to time on this list, but since 
there are new people joining all the time, it bears repeating:

User interface design is hard. HARD. It is a result of compromise 
between conflicting goals. You can never please a random group of 10 
people, much less everyone.

Since I do Paul's panel mockups, I get the privilege of injecting my 
opinions about the interface of new modules sometimes. Usually that 
consists of making sure that similar functions on different modules 
are called the same thing whenever possible. In the specific case 
mentioned with the 'V/R' switches on the MOTM-190, I agree - the 
minimal labels are less than ideal. We went through 4 iterations, 
seeking list feedback along the way, to get to the final design. I 
think it's interesting to review how the process went:

Take one: switches were in the current position, and were 
labelled 'VCA/RM' and 'LIN/EXP'. The top knob was labelled 'MODE', 
with the left extremity labelled 'AM' and the right extremity 
labelled 'RM'. This one didn't work because the labelling was just 
too crowded around the left border between the switch and the pot 
graphics.

Take two: switches were moved underneath the bottom pot. This looked 
good on paper, but in reality the switches were too close to the 
jacks to comfortably grab when a plug was present.

Take three: move the switches back, shorten the labels to 'V/R' 
and 'L/E'. Label density was better, but the top knob's function 
still was a bit cryptic.

Take four (final): top knob was relabelled 'BLEND', and the 
extremities were relabelled 'IN' and 'RM'.

One problem with layout on this module is that Paul is cramming a 
lot more functionality and features into 1U than ever before. Given 
the tradeoff of space for labelling clarity, I like the choice he 
made. I'd rather be able to fit twice as many VCAs in my scarce cab 
space and learn what the switch labels do, than to go to 2U and have 
better switch labels. Others may not like that decision but  
eventually you have to make a choice and just go with something.

Here are some of the constraints one must consider when designing 
for MOTM: some of these constraints are valid with other formats 
also.

1. fixed height (naturally!)
2. jack field at the bottom
3. standard placement grid
4. can't always use all available panel space if pcb collides
5. most pots need to be on the far right column because that's where 
the pcb is
6. cramming the features in the module space allotted

If you violate any of the above, it either becomes impractical to 
build, or people complain because a module doesn't have the uniform 
look that they expect (example - the pot spacing on the MOTM-450). 
However, you HAVE to allow for some flexibility in design, because 
sometimes some modules just don't lend themselves to a certain 
format. 

FWIW, the most useable format ever achieved IMHO is the E-MU (old 
timer list members groan in unison: 'there he goes again!') I like 
the E-MU scheme of audio inputs to the left, control inputs at the 
bottom, and outputs to the right. But, it's a total space hog! I 
find the MOTM format to be the best compromise between ergonomics 
and functional density. The fact that at least 3 other manufacturers 
build modules that comply with this format speaks for its 
effectiveness and desirability.

Do I think the format could be enhanced? Sure! Here's a few things I 
can think of:

1. Be flexible with the MOTM grid when necessary. The MOTM-450 is 
the first 'official' module to do this, but many of my sequencer 
designs also necessitate this. I'm not saying to break the grid for 
trivial reasons, but if there is a complex module with a compelling 
reason...

2. I like the idea of different knobs for different functions, when 
it makes sense. Examples include chicken beak pointer knobs for 
rotary switches (JLH-822), and small knobs (the very dense UEG). I'd 
like to see some experimentation with mixed small and large knob 
sizes in the same module - in a dense module, small knobs could be 
used for less important adjustments, while the major functions have 
full sized or even oversized knobs. Look at JH's JH-5 design, which 
uses 4 knob sizes to great effect.

3. Module edge markings. Sigh. MOTM's Achille's heel. There's no use 
lobbying Paul to change now - the visual scheme has been well 
established. I swear that before I die, I'll figure out an elegant 
way to add these after the fact - something better than hand taping!

In spite of these minor issues, I still think that it's a very easy 
and intuitive interface to use.

Enough ranting from me for now.

Moe

http://www.hotrodmotm.com

> ----- Original Message ----- 
> From: mate_stubb <mate_stubb@...>
> We have this discussion from time to time on this list,
> but since there are new people joining all the time, it
> bears repeating:
------snip-------

all great insight Moe... until I got to this part:
now ROTFLMAO.....

> I'd rather be able to fit twice as many VCAs in my 
> scarce cab space ....  --- snip -----

SuperMoe cabinet 1+2 = 4 rows 18U each x 2 = 144U wide
Scarce?  What a Stooge <snicker>

Larry (the other Stooge)

I'm really encouraged by Moe's e-mail on future improvements to user 
interface.  The compromises imposed on the user interface by the 
conformity to a "grid" of knobs are, IMHO, a source of many of the 
complaints.  And allowing different knob sizes will be a big gain.  
For example, the micro VCA's labeling could have been more complete 
with the extra real estate gained by using some smaller knobs for 
less frequently used functions.

Extra functions in modules are of course welcome, but there is a 
point at which the usability of an interface is hampered by the 
compromises imposed by having EVERY function presented at the same 
level of importance to the user.  For example, if there are 5 
interesting but arcane functions that are rarely used on a module 
that crowd the one often used function, making it hard to find within 
the grid of identical knobs, the module is compromised.  Why not have 
a normal knob with some space around it for the often used function, 
and smaller knobs (to the side or below) for those arcane functions 
we all secretly love?  I hope this will be possible in future MOTM 
modules.  I'd gladly give up F-111 washers for this!

As an aside, the latest BMW 7-series sedan incorporates a new CPU-
based dash, with menus and pseudo joystick to access dozens of 
functions.  Many people hate it.  Like the DX7, too many buried 
functions.  A modular CAN fall into the same trap, even though all 
functions are accessable simultaneously, if the functions are buried 
as needles in haystacks of identical knobs, spaced identically in 
grids with limited labeling.  Dare I say it... it reminds me of 
searching through menus...

Best regards,
Roger

The issue is one of stocking.

I have to deal with Tyco for the knobs. You know, the company that the CEO is in the slammer for
tax evasion and the $6,000 shower curtain.

Nothing prevents you from using your own knobs.

I will be the FIRST TO ADMIT that MOTM is plain. Boring. Simple. Kind of like me :)

Sorry! The panels are expensive enough without 4-color anodized graphics.

Paul S.

> For example, if there are 5
> interesting but arcane functions that are rarely used on a module
> that crowd the one often used function, making it hard to find within
> the grid of identical knobs, the module is compromised.

But this implies that you have to get people to agree on which functions are
arcane and which aren't. It has been shown that it is difficult to form a
consensus of opinion even when all of the functions have equal importance,
e.g., the layout of the filter bank's knobs. I submit that it would be even
more difficult to get people to agree on which functions in a given module
are more important than others. If I consider my own (relatively brief) use
of my MOTM system, it is not clear to me that I would consider any of the
functions on some of the more knob-laden modules--say the 300 or the
filters--to be subordinate in importance to others. It all depends on how
I'm using the module on a given occasion. Maybe one day I'm not using the FM
inputs at all; maybe the next day the adjustment of the FM levels is
critical to the operation of the patch.

With a modular synthesizer, to make decisions about which function is more
important than another means to predetermine in the mind of the user which
functions are more important than others. It is my opinion--and yes, it's
just an opinion--that this design philosophy would be flawed. A modular
synthesizer is like a blank canvas and a set of paints. To tell the user
that one function is more important than another would be like telling the
painter that red is more important than blue.

--Adam

Ahem...

Let it be said, in the land of MOTM, green is more important than either red
or blue. :)

But seriously (sort of), I agree with Adam that the features of a module
that are in use for a given patch are the most important at that time, so
it's difficult to say which features are more important than others.  I also
agree that there's NO way you're going to get a consensus on
functions/interfaces/layouts, so why bother trying?  I like the common theme
for MOTM.  The painter's palette analogy is a good one.

Conversely, I think there is merit to having some differentiation between
the various controls in a module, be it knob size, shape or position, and
between modules as well.  This aids in pattern recognition of functions, and
therefore, learning and ease of use.  For example, a larger Fc knob on a
filter, or chicken beak knobs on Larry's excellent 822. <plug>

As far as crammed functions into small areas, well, one of the [many]
reasons I chose MOTM was it's "comfortable" utilitarian layout.  There's
plenty of space.  Black + White = Perfect.  Alas, I think the 190 is the 1st
module to violate the comfort index.  But I understand that Paul needs a 1U
VCA, and I appreciate the extra functions it sports - although I would
rather have it in 2U.  I guess we'll get a 2U VCA when the *cough* 130
*cough* comes to fruition. (*cough*)

Oh, and by the way, I think the more arcane functions, the better.  Bring on
the 500 series!  Please!

George - just full of opinions today

----- Original Message -----
From: Adam Schabtach <adam@...>
To: <motm@yahoogroups.com>
Sent: Monday, August 12, 2002 1:07 PM
Subject: Re: [motm] Re: User Interface design - Moe's long semiannual rant


> > For example, if there are 5
> > interesting but arcane functions that are rarely used on a module
> > that crowd the one often used function, making it hard to find within
> > the grid of identical knobs, the module is compromised.
>
> But this implies that you have to get people to agree on which functions
are
> arcane and which aren't. It has been shown that it is difficult to form a
> consensus of opinion even when all of the functions have equal importance,
> e.g., the layout of the filter bank's knobs. I submit that it would be
even
> more difficult to get people to agree on which functions in a given module
> are more important than others. If I consider my own (relatively brief)
use
> of my MOTM system, it is not clear to me that I would consider any of the
> functions on some of the more knob-laden modules--say the 300 or the
> filters--to be subordinate in importance to others. It all depends on how
> I'm using the module on a given occasion. Maybe one day I'm not using the
FM

> inputs at all; maybe the next day the adjustment of the FM levels is
> critical to the operation of the patch.
>
> With a modular synthesizer, to make decisions about which function is more
> important than another means to predetermine in the mind of the user which
> functions are more important than others. It is my opinion--and yes, it's
> just an opinion--that this design philosophy would be flawed. A modular
> synthesizer is like a blank canvas and a set of paints. To tell the user
> that one function is more important than another would be like telling the
> painter that red is more important than blue.
>
> --Adam

Hi,

In a message dated 8/12/02 12:27:51 AM Central Daylight Time, 
mate_stubb@... writes:

> 2. I like the idea of different knobs for different functions, when 
>  it makes sense. Examples include chicken beak pointer knobs for 
>  rotary switches (JLH-822), and small knobs (the very dense UEG).


 I like the idea of different size knobs/shapes for different functions.  Any 
smaller than the UEG knob size would be too small though, IMO. I wonder what 
smaller sized knobs would do to the layout? To make it mesh visually, would 
the pot spacing have to be changed for a module that had groupings of smaller 
knobs?


  
>  3. Module edge markings. Sigh. MOTM's Achille's heel. 

  What about Stooge panels made with the edge markings? Maybe a couple of 
thoughtfully-placed modules with edge marked Stooge panels would help with 
this.

 As far as other ideas... Since the 2U modules often have mixer inputs with 
matching level pots, how about an outline (like the switches have) around the 
level pots? Come to think of it, how about an outline around all pots that 
have similar functions on the 2U modules? Looking at my 420 filter, possible 
knob outline groupings could be [Freq / Res], [FM1 / FM2], [In1 / In2 / In3].


 (anxiously awaiting my newly-placed MOTM order. :P )
 Andrew

Now there's an idea!  DIY knob customization.  With 20 some modules out
this is better than asking for the module layouts to change.
John Loffink
jloffink@... 

Nothing prevents you from using your own knobs.

> Now there’s an idea! DIY knob customization. With 20 some modules out 
> this is better than asking for the module layouts to change.

Yes, a change in panel design now wouldn't do much good for those of us 
who have already built systems. Which is not to say that this is reason 
enough to not pursue improvements in panel design, but to point out 
that any such improvements would largely be lost on those already 
accustomed to the current design.

But yes, customized knobs are a wide-open opportunity. I was thinking 
earlier today that one could differentiate functions by coloring the 
silver plates on the knobs differently, in addition to (or instead of) 
using different-sized knobs. You could make a tacky attempt at such a 
scheme by using different colors of Sharpie markers. A slightly better 
technique would be those round colored stickers available at office 
supply stores. Another less cheap-looking possibility is the 
self-adhesive color film used by graphic artists. It's translucent, so 
it might allow some of the metallic quality of the silver plates to 
show through--looking like a sort of faux anodization, perhaps.

Then one could come up with whatever color scheme made sense, given 
ones inclination. Maybe all frequency-related knobs are one color and 
modulation levels are another, or maybe all oscillators have the same 
color knobs, or whatever.

--Adam

> Conversely, I think there is merit to having some differentiation 
> between
> the various controls in a module, be it knob size, shape or position, 
> and
> between modules as well.  This aids in pattern recognition of 
> functions, and
> therefore, learning and ease of use.  For example, a larger Fc knob on 
> a
> filter, or chicken beak knobs on Larry's excellent 822. <plug>

I'll buy that. I can't see that one could establish a ranking of 
importance of parameters, but I agree that differentiation of 
parameters by knob variations could be helpful. Or maybe it comes down 
to the same thing: someone might differentiate the knobs according to 
their ranking of importance, but to me it would just look like 
arbitrary differentiation. :-)

--Adam

> >  3. Module edge markings. Sigh. MOTM's Achille's heel. 

I'm sitting here on my girlfriend's computer, without any access to a 
visual program like photoshop, but i had an idea for edge markings:

Think relatively thin, transparent plastic.

Think shaped like an "I"

SO (hoping we're all monospaced here):


XXX|XXX
X X|X X
XXX|XXX
   |
   |
   |
   |
   |
   |
   |
   |
   |
   |
   |
XXX|XXX
X X|X X
XXX|XXX

What you have are tabs on the top and bottom of a transparent piece 
of plastic.  These tabs have holes in them that are aligned with the 
edge screw holes of two modules.  a thin (1/8"?  1/4"?) strip runs 
down the middle, with...

a white line silkscreened on it, slightly wider than the gap you 
might have between two modules given the slop around the mounting 
holes.

comments?  requests for clarification?

I have thought long and hard, and there are three schemes I can come 
up with:

1. Paint it. Unfortunately permanent.
2. Plastic or metal overlay (your exact idea)
3. Plastic or metal edge channel

Your idea would work, but somebody would have to manufacture the 
stuff. Also, I'd have to see it actually on some modules to decide if 
the mounting tabs are unsightly. One bonus is that the tabs could 
help prevent rack rash.

For edge channel, the problem is finding stuff thin enough. You can 
buy brass or aluminum 1/8" channel, but it's too thick to keep the 
modules on the MOTM rail grid.

Lately I've been thinking about a thin foil, with adhesive on one 
side. Anybody know where I could get a material like this?

Moe

>>>>
Think relatively thin, transparent plastic.

Think shaped like an "I"
<<<<

> Lately I've been thinking about a thin foil, with adhesive on one
> side. Anybody know where I could get a material like this?

I bet you could find it at a well-stocked craft store. You know, the sort of
place that sells picture frames, fabric paints, needlepoint supplies,
plastic flowers, that kind of stuff. (I'm always tempted by the faux stone
and marble finishes these places sell. I think marble front panels would be
cool.)

Many years ago I had some narrow, stick-on vinyl striping which was sold for
automotive pin-striping. (I used it to decorate my skateboard helmet.) It
might work also if it came in colors. The stuff I had was black, which
wouldn't help much.

--Adam

> > Lately I've been thinking about a thin foil, with adhesive on one
> > side. Anybody know where I could get a material like this?

This should be a hardware store item, Dave.  Like copper tape I'm 
planning to use for my ribbon controller, you could find aluminum 
tape (also known as racer tape or 200mph tape) and brass tape.  Color 
choice is limited. Then Adam wrote: 
> Many years ago I had some narrow, stick-on vinyl striping which was 
sold for
> automotive pin-striping. (I used it to decorate my skateboard 
helmet.) It
> might work also if it came in colors. The stuff I had was black, 
which
> wouldn't help much.
 
Ah, but it DOES come in MANY colors!  Check out your local auto body 
supply or auto paint shop, hardware store, gee-I'll bet Kmart or 
Walmart in the automotive accessories aisle, auto  parts store (Pep 
Boys, Champion, Big A, Auto Zone, etc...) or even some big service 
stations.

Not willing to put on pants and peruse the automotive delights of 
Ames, Dave?  Go mail order- J.C Whitney has been putting out gobs of 
auto parts and accessories for years.  They've finally(!) stopped 
sending me catalogs and are on the web at http://www.jcwhitney.com/.

The pinstriping tape may well be the best idea-it doesn't shrink and 
has an aggressive adhesive.  With a little creativity, you could make 
your MOTM look like a "Von Dutch" original!  Bitchin' dude!

Chub

So, as often, Yahoo has mangled ascii art.  Apparently sometimes it 
shows up fine.  the line down the middle is supposed to be down the 
middle.  If it's down the edge on your machine, then CURSE YAHOO!  
Heh.

--- In motm@y..., "sucrosemusic" <sucrosemusic@y...> wrote:

> > >  3. Module edge markings. Sigh. MOTM's Achille's heel. 
> 
> I'm sitting here on my girlfriend's computer, without any access to 
> 
> 
> XXX|XXX
> X X|X X
> XXX|XXX
>    |
>    |