The New Google Logo and Kerning | Hacker News Reader

The New Google Logo and Kerning

ak.vbroek.org

56 pointsmillisecond47 comments

It's not like the Google designers didn't try to - alteast partially - mitgate the perceived space. For whatever reason they didn't.

But, atleast in my humble opinion, her kerning is way better.The 'gle' doesn't seem so squished, but the 'G' shouldn't use the same space as between the 'o's. It's just way bigger (vertically and horizontally) and more "qubic". So I would sggest to add 1-2px between 'G' and 'o' and it would be even better

No insult to her, but I think Google's looks better. It just looks cleaner for some reason.

Both 'felt' differently good. Google tighter one feels a tiny bit oppressing now that I've read the article. Free that 'l'.

btw: http://imgur.com/sasrZ6F the favicon is delightfully sunshiny. I kept looking at it for the first week post-redesign.

Oh yeah it's super nice.

I must admit like google's version better... That space between the O and L looks waaaay too far out to me. I guess it's pretty subjective.

Yeah, from the moment I saw the new logo I seemed to have an instinctive reaction against the kerning, it just feels wrong. Her spacing is a bit better, but is still by no means perfect. With a bit more iteration I'm sure it could be worked into something the vast majority accept...

... which leads into the question of why Google didn't take the time to address the issue properly? It seems that if you're going to change the logo of a bazillion-dollar company you'd make the effort to get it right. You, me and the OP can't be the only people in the world who disliked the kerning, so it's not as if the Google designers weren't aware it was an issue (unless they completely ignored testing it?) Just seems a bit shoddy.

I feel Google's kerning is too squished, but hers to too far apart.

I wonder if one could come up with some kind of formula to calculate the distance between the letters as perceived by the majority of people, based on say pixel density.

Strange, I really felt like Google's kerning was too far apart.

That sounds like the kind of a data-driven, "42 shades of blue" approach Google is known for. :) There was a kerning game on Hacker News a while ago. The player's goal was to arrange the letters of a word to the "correct" kerning. Now imagine something like that at Google scale, like captchas or the Google Image tag matchin game.

It's possible that "good kerning" interacts with factors like distance from screen and font size. So you cannot isomorphically scale up. That would mean Google would actually need different kerning for different contexts.

"I feel that the intensity of the childlike design grates against the realities of Google as a company — which generally I am a big fan of but — that involve some pretty weighty and adult issues involving privacy and user rights. I don’t want Google to be childlike. Fun, yes. Childlike, no. So, the design feels optimistically naive or pessimistically deceptive."

I have no idea how we went from a discussion about kerning to a discussion regarding the nature of Google's corporate structure and values.

Because design is not about making things pretty, but about making things attractive. Google is not changing their design just because. Behind the childish fonts, colorful interfaces and simple icons there is a careful thought about what Google believes it's the best way to present itself to its costumers.

How Google is, or more specifically, how Google wants people to think it is, it's implicitly present in every single design choice, even in a seemingly irrelevant one as the relative distribution of the logo's letters.

And for a company of Google's size, it's safe to say Google was considering the same issues OP was too, and what their branding conveyed regarding them. Yep, even down to the pixel.

I'll admit i don't perceive much difference. But:

> Of these, most prominently, is the desire to appear playful and childlike. I see it.... However, more strongly, I feel that the intensity of the childlike design grates against the realities of Google as a company—which generally I am a big fan of but—that involve some pretty weighty and adult issues involving privacy and user rights. I don’t want Google to be childlike.

Yes, that's exactly what Google DOESN'T want you to think about, which is perhaps why they want to appear playful and childlike.

Of course Angela's kerning is 100% better, but in terms of branding Google has always been "anti-design" so maybe the bad kerning was planned?

Anti-design, but they A-B tested different shades of blue to get it "just right"? ;)

Maybe "against the grain" or "intentionally quirky" is more accurate?

I thought that was Marissa at Yahoo?

It was Marissa at Google. 30 Yahoo! logos in 30 days was Marissa at Yahoo!.

I actually like Google's kerning better. In Angela's version, the l stands out too much. Its unique green colour already makes it more noticeable than the other letters and spacing it out this much makes it too prominent.

Yeah, I think she overdid the kerning on the "l". It looks separated from the rest of the word. Perhaps 13px (compromise between Google's 11px version and Angela's 15px version) would work better?

Speaking of color, the yellow "o" blends into the white background and makes the space between the two "o"s, as well as the space between "o" and "g", look smaller than they actually are. In order to look consistent, the other characters need to be closer together than they would ideally be in a black-and-white version.

Google's kerning appears to be perfectly logical, mathematically speaking, according to the rules pointed out in the article. Rounded letters have a larger perceived space and so they need less kerning space.

The difference seems to come down to whether half-round letters like the G are treated the same as the fully rounded letters like the two o's.

To my eyes they look a bit different, but I can't say that one is any better than the other.

What? It's subjective. How is it 100% better?

How much you kern may be subjective, but kerning itself (and other aspects of typography) are very related to legibility which is less subjective. Of course on a one word logo that really won't make a measurable difference.

I'm pretty sure google isn't "anti-design"

I assume OP was mentioning that Google is anti current design trends

I can't tell the difference, honestly.

So there's this phenomenon that I see on occasion that I call "Invented problems" where somebody invents some concept like a 'soul' just so they can sell you their authority in saving it.

I'm not trying to belittle anyone for being a kerning snob, since I know there's a good amount of pleasure and life to be had in concerning yourself with a narrow passion, but I can't shake the feeling that people who care about kerning more than the point of simply knowing that it exists and is called that as a matter of trivia haven't been sold an invented problem. And also that the efforts they put into it are so myopic that it risks tone-deafness.

For example, I keep hearing that monospaced fonts are 'bad' because "m's are crunched together, but i's have all this room..." Ok... That adds character to your words. The sequence "iiii" is loose and free, and the sequence "mmmm" is busy and monolithic. Why is that a problem? It 'hurts readability' somehow? That's like saying that personality hurts relationships because you can't just put everyone in the same group. The words 'phenomenon' and 'somebody' feel different from 'soul' and 'authority' because they have different shapes, and that might be a very good thing.

Anyway, my point is this: whenever I read about somebody putting significant effort into something like kerning, I'm left feeling like they don't even know if they are making a huge mistake and 'fixing' something that was already optimal, and in fact, could be optimal without anyone giving a single real thought on the kerning. All this effort is being spent on something that isn't actually making the world better in any relevant way.

Kerning isn't an invented problem like saving your soul. Kerning exists whether you believe in it or not.

It's a core aspect of all typography, almost as much as the letter shapes themselves. The real problem is that like many disciplines that tightly intertwine design and engineering aspects, the right answers aren't quite as precise or objective as you might hope.

Kerning isn't the invented problem. The importance of marginal gains by adjusting kerning in logo design is.

Imagine if I wrote an article where I was claiming that the blue component of the logo color should be 3% lighter. Maybe it's a real improvement. My concern is that the color of the logo on that scale is such a small part of what makes it a good design or not, that my reflex of analyzing it in that way is counter-productive to being a good designer. I'm not saying color is unimportant. I'm saying the focus might be too narrow to be used to do design.

This post is beautiful -- it perfectly describes itself.

Design absolutely makes the world better in a relevant way, your day to day existence would be dramatically more painful if everyone throughout history thought the way you do. The fact that you do not care to understand it does not make it less relevant or beneficial to you. You are like the anarchist complaining about the utter uselessness of government all the while driving on public roads and drinking public water.

When you have the urge to go on a rant like this you would do well to stop and ask yourself, "is there something here I just don't understand yet?" and save yourself some embarassment.

I'm not embarrassed about what I said. Maybe you think I should be, and I put extra effort into saying it because I knew it would be a controversial issue.

But I wasn't talking about design in general, I was talking only about kerning. I simply can't imagine that the kerning of text that I read has made my life less painful in any way, except in the situation where we've avoided clearly unreadable, awful kerning. Now tell me how the Google logo is that kind of situation, and maybe I'd understand your point.

Also, I do care to understand this subject. I just also care to understand more than just that. We can talk all day about kerning, but you don't do good design by talking all day about kerning, you do it by talking about human experiences. And as a human experience, kerning isn't really on most people's radar.

You put effort into saying that? Because it read like an undifferentiated rant against all typography.

If it was only about kerning, why did you set up a huge strawman about monospaced fonts implying that you would by fine in a world with only monospaced fonts, and that readability is not really a thing, and then going on to set up some analogy about relationships that is utterly unfathomable in its irrelevance. If it's just about kerning then I suppose you didn't intend any of this, but that's definitely how it comes off.

When I was talking about monospaced fonts I was doing so only to point out that people tend to talk about abolishing them like it's all upside. My point is to suggest that one take a broader view of the subject before advocating a design position, and focusing on the kerning of the Google logo is an example of failing to do that.

If I had written more, it would probably look more like a rant, and if I'd written less, it would've been bull-in-china-shop writing. Writing subtle criticism on the internet is walking a fine line.

In addition to letter shape, the letter colors (in relation to each other) also change the perceived whitespace. So, in my opinion, it should be kerned differently than her proposal, and Google's version looks better to me. But, frankly, the yellow "o" will always make it look off to me.

I prefer the slightly closer kerning of the Goog in her version, however I much prefer the current kerning for the l.

Having said that, I don't think there's anything wrong with the current kerning. I actually suspect the wide spacing is more do with readability when scaled down more than anything else.

Kerning and tracking are often adjusted differently at different point sizes, especially if you are designing logos and can easily tweak them manually.

Absolutely, and for most logos that's fine, however when you're the size of Google, controlling which version of your logo someone decides to use isn't easy. Sometimes it's better to assume that people are going to use your logo at sizes you didn't intend and in places you never expected, so having a bit of breathing room isn't a bad idea.

Having said that, I don't really expect Google would have chosen their current kerning simply so that it (potentially) downscales better - I'm sure they have more important criteria.

Tim Gunn, not "Time Gunn," but—fwiw—if I was starting a band, I would totally name it "Time Gunn." Make it work :)

Great eye and I have to say, it really makes me think twice now that I look at the logo every time. However, it's still a subjective matter. Now the question is how long did it take the Google design team to decide on the Kerning? Must have been a vote imo.

If you read through the comments there's some interesting info and background including that the kerning of the new logo is not consistent on their website and the one on the homepage is "more off" than other versions. Also a link to a Russian designer's take on the same issue.

Would be interesting to see some data on how people perceive the relative kerning between curved letters as opposed to straight lines.

"Each time I pulled up the search engine (which is often), I spent more time than usual staring at the doodle above the search bar."

Does this person actually type "www.google.com" into the address bar and press enter?

Maybe it's editorial license and that's not actually the case, but it's still concerning.

Maybe Angela just has her homepage/new tab set to "www.google.com" before she performs searches :) Some people like to see the google doodles.

Maybe that's her new tab default page.

On Chrome, it displays the Google logo on every New Tab page (by default)

I do that. Why does it concern you?

Reply on news.ycombinator.com