Draconaei’s Blog

If anyone was curious about the final 11 designs mentioned in the previous post, voila!  Here they are.

Exploring Good Logo Design.

A necessary disclaimer: I love scientists.  I work with some of the most brilliant minds in the field of high-energy-density laboratory astrophysics (say that ten times fast!)  In fact, I would like to think of myself as well-rounded, and being both a scientist and an artist.  That being said, let’s use this opportunity to explore good logo design through some oft-made mistakes, as highlighted by my experience designing a logo for CRASH.

C.R.A.S.H. is the Center for Radiative Shock Hydrodynamics.  This team of scientists focuses on developing a code to accurately simulate radiative shocks: energy moving significantly faster than the speed of sound, which simultaneously releases energy in the form of radiation.  The professor who runs my lab also heads the CRASH team, and knowing I dabble in graphic design, he asked me to design their logo.   There the learning experience began.

Mistake #1: Not understanding what a logo really is.

A logo is a simple, clean image whose design should be unique, appealing, and meaningfully represent an idea- but it should be just that.  A representation.  Problems arise when a logo is either taken too literally, or has no real connection to what it represents.

• When presenting a potential logo with a radiative shock facing to the left, an exasperated scientist spoke up.  “But our simulations display shocks moving to the right!  If we chose this logo, we would have to flip ALL of our data to match it!”
• Another design included a data point with error bars, and received a similar response: “REAL data includes at least three data points, presumably improving over time.  We recommend three circles, each with an error bar, decreasing in size from left to right.”

Mistake #2: Understanding concepts, but not how they graphically translate.

Logo designs should be simple, and scale down to small sizes without losing their overall effect; because of this, they can’t include too much information, but should still include enough to be recognizable.  This requires knowing very specifically what information you want to include, and filtering out irrelevant or overtly complicated concepts.

• After some discussion, the CRASH team agreed that the logo should best represent “uncertainty quantification, ” otherwise stated as, the study of accurately measuring potential error from specific experimental parameters.  I challenge ANY of you to do this.
• The above request was followed by, “We would prefer the design to be very simple.”

Mistake #3: Overloading a design with “cool” effects- i.e. gradients, shadows, and reflections.

Designers and non-designers alike can be gripped with the temptation to follow the latest design trends, which often include extraneous fancy graphic effects.  A good logo design could use these elements, but should never depend on them for effectiveness.  The design should work well even when reduced to stencil or greyscale format.

Assuming you agree with the above, the following suggestions should speak for themselves:

• In reference to a maize and blue color scheme:
  • “Add maroon!”
  • “The shock does not look radiative enough.  Please make it red.”
  • “You should add several more colors.  There are more colors in the rainbow, you know.”

Another way to say this is, if the logo doesn’t look good already, special effects won’t fix it.

Mistake #4: Trying to incorporate every persons’ ideas and suggestions into a single logo design.

Including everyone’s ideas is a noble cause, but in the end, not everyone can get the design they want, especially if the group is especially disagreeable.  The important thing is satisfy the group’s basic needs, with a design that will retain its effectiveness over time.

• After 50+ design iterations, I emailed the CRASH team a selection of 11 logos, and asked everyone to vote.  The response was an almost perfectly even distribution; if I had sensed conflicting interests within the group previously, I now had solid confirmation.
• At the very least, it would be impossible to realistically include all three of the following ideas in a single logo, each idea being suggested by a different member:
  • “Design a shockwave moving through the word CRASH.”
  • “Why don’t we just take a snapshot of our data and add text to it?”
  • “You should include a car.”

This concludes our discussion on logo design!  So as not to leave a cliffhanger, here is the design that was chosen in the end- a slightly stylized graphic representation of simulation output.