Greg McKaskle [EMAIL PROTECTED] said:
>While it is always good to get feedback, it is even better to get
>feedback that you understand.
I apologise if my comment was provocative - it is only an opinion, but it wouldn't be
the first time someone on info-labview has had one, and I'm glad to elaborate on mine:
>Some of them are available with alternatives and at a couple sizes, but
>I didn't see twelve XY graphs. Are we looking at the same page?
By 'XY graph' I meant a representation of a signal or curve on Cartesian axes - I
should have been clearer, because most of the examples I had in mind are actually
signals plotted against time. Looking at just the letters A-F I can see stylistic
differences between each of the following: amplitude, AM (options 1 and 2), analog,
analog input, autorange, channel, clear all arbitrary waveforms, configure, digital
data, digital waveform, frequency (options 1 and 2), and frequency generator - for
example, the presence or absence and colour of the axes, the colour of the graph line,
the presence or absence of tick marks, the presence or absence of arrows on the axes,
and the style of measuring bar used to indicate the amplitude or period of a signal. I
don't see any consistent meaning to these stylistic differences, for example why the
axes are sometimes grey, sometimes black and on one occasion red - it just seems to be
random, and that's what I mean by a 'haphazard ragbag'. 'Hideous' you a
I'm an interested amateur rather than an expert in graphic design and the psychology
of human-computer interfaces, but I believe there are two philosophies to choose from
in creating an iconic symbol: to try and explain a meaning by illustrating a concept
("let's see, that's got the symbol for erasing and the symbol for a waveform, so it
must mean 'erase the waveform'") or to be distinctive and recognisable without
necessarily being illustrative ("oh yeah, the one with the eye, that's Photoshop").
It seems to me the symbols here are falling between two stools because there's no
*consistent* symbol for 'waveform' to support the first mode of perception, but there
are also few eyecatching visual concepts that exploit the second (the hourglass for
'delay', jugs for max/min, stethoscope for 'probe' and penknife for 'utility' might be
among the exceptions here). If colours and styles vary at random, or at the artist's
whim, it means first of all that the viewer wastes time working out that this square
wave means the same thing as that square wave, or that those graph ticks don't mean
anything special whereas the ticks on that ruler do, and second that opportunities for
conveying extra meaning are being lost - perhaps that a red waveform is always an
output and a blue one an input, etc. There are some individual symbols I'd take issue
with (do you really 'abort' when you get to a stop light, and if so how do you make it
to work every day?) but mostly the problem is one of applying the
I'm not saying that this is easy to do well, and consistency by itself doesn't
guarantee understandability (hands up who can instantly name all the LabVIEW string
functions from their icons?), but I'm a passionate fan of the graphical paradigm of
LabVIEW and I think it can be reinforced or undermined by good or bad icon design. I
think it'd be a shame if some of the (IMHO) flaws in this collection of symbols became
established as a standard.
I'd be happy to pass on these comments to the appropriate person at NI and to offer
more feedback if it's desired - would the 'product feedback' or 'ni.com feedback' form
be appropriate?
Tom
Dr Tom Hawkins
High Throughput Screening Analyst, Associated Octel ltd
PO Box 17, Ellesmere Port, Cheshire CH65 4HF, UK
+44 (0)151 355 3611 [EMAIL PROTECTED]
-----------------------------------------
Email provided by http://www.ntlhome.com/
