[ebbv]

I think there's some valid points in this article, but there's also a valid point that kind of undermines a central thesis to this article:

A logo does not need to tell you what it's for if your first viewing is out of context. In fact, 99% of good logos won't accomplish that task. The point is to be recognizable once you know what it is already, and be pleasant to look at over and over.

Now, whether any of the minimalist logos meet that criteria is up to the user. But judging logos on whether you can tell what they're for out of context is a false criteria.

How would you know the Apple logo is for a computer company out of context? You'd be more apt to guess it was for an orchard or cider company. How would you know the modern Windows logo is for the operating system out of context, and not just a bunch of boxes for a moving company or something.

Ambiguity in logos is not remotely a bad thing. What makes a good logo or bad logo is mostly subjective.

The main objective thing I would damn most of the logos highlighted in the article on is being so derivative that they become tedious.

[sloreti]

> What makes a good logo or bad logo is mostly subjective.

But not entirely. There are some choices that can inherently weaken a design. A logo should be scalable; it should look just as good on a business card as it does on the side of a plane. A color logo should be able to be translated to black & white.

[TeMPOraL]

True. But most of those rules can be taught over a single lecture. Why then the need for expensive designers? I suspect most of the philosophy around logo design is a simple job-security measure, whether they admit it to themselves or not. For a logo, a person with some sense of aesthetics and knowledge of aforementioned rules is more than sufficient.