I'm not complaining, the technology is cool and no doubt it will get to a point that it can critique objectively a logo, but it's not yet here.
1: https://www.reddit.com/r/Design/comments/6wz4ho/logo_rank_is...
I'm not complaining, the technology is cool and no doubt it will get to a point that it can critique objectively a logo, but it's not yet here.
1: https://www.reddit.com/r/Design/comments/6wz4ho/logo_rank_is...
Is the Nike logo really a unique logo purely from a design perspective, or has its extremely strong brand made it a successful logo despite its actual design not being very unique? Whereas a dumb meme is the opposite -- terrible brand, but very unique.
The tool clearly can only look at actual visual design, not branding or any meaning within the image. And in that limited context, it seems successful. I'm not sure it is reasonable to expect it to go beyond a purely visual evaluation of the logo.
- on a uniqueness front there are a lot of swooshes, peppers and moustaches that look sort of similar to nike, whereas that image is guaranteed to not have been used for any icon or logo, ever.
- both are fairly legible
- the meme is certainly more colorful and eye-catching
the neural net doesn't really judge the aesthetics of a logo, and it doesn't have any idea what the content of the image is. It just impassively gives you a score based on the similarity to thenounproject icons and a few other parameters.
My point is that your tool is very cool, I wish I could do something like that—but that's not a «logo rank», it's a «how unique, legible and eye-catching the image you upload is» rank.