> We found that what we liked best in a Müller-Brockmann book or on the screen in photoshop was not what we liked best on real data. Photoshop type rendering is pretty poor, which is another reason going straight from print to code worked well for us.
Really good call to omit Photoshop altogether from design endeavors like this. It's a waste of time copy/pasting actual data into your comp, whenever you want to change part of your design, you have to repeat it manually with all the other elements on the page, the fonts render poorly...
We lately try to go from sketch to simple HTML mockups. Having actual data readily available while designing is such an eye-opener. Thanks to web inspector, we can tweak the styles and see the changes get reflected in the entire page. Sometimes when we want to play with the layout, we take a screen grab, cut and move things around in Fireworks.
Nicholas Felton, about the Facebook timeline design:
"You can come into a meeting with a very beautiful comp and it’s like, ‘Oh yes, we should do it that way,’. But you’re never going to know if you can do it that way until you pump in the real data and live with it for days or weeks".
I don't understand how Photoshop is still considered a web design tool. Unless designing a marketing site with no content, it hinders good design more than it facilitates.
And oddly enough, they probably would have gotten at least closer to right than to wrong.
I love discussions like this though, because it tells me the 'why' and 'how' somebody else got to their decision. I am not well-versed on typography. While I believe I have fairly high standards for font selection, I honestly couldn't tell you whether or not one font should be preferred over a similar font.
Articles like this help to at least enlighten folks like myself into knowing the criteria. In much the same way that a person who has little experience with cars might not know to ask about miles per gallon, safety ratings, etc., this kind of information lets people know that there's more to care about than 'leather vs cloth interior'.
I'm not saying that details aren't important. But I think there are more important problems for them to figure out now like "how are we going to determine what people are interested in by mining Tweets and Facebook posts. These information could be totally fallacious."
The higher level product design problems you're talking about are super important, and we'll write articles about that soon. We're really into interaction design.
We're not figuring these problems out now. Our user research started over a year ago, and we've iterated with many different versions of the product ever since. Our engagement numbers are fantastic.
However, being smart about product design and optimizing engagement does not preclude the attention to detail we demonstrate in this post on type.
The role of type in a content focused design is just a worthy topic to post on, and there's no more to read into it than that.
1. The homepage really doesn't prove the attention to detail thing. It can be improved a lot. By the time someone wrote this blog post, he could have fixed that. (I really liked reading the post, do not get me wrong.)
2. Aesthetics is important. But would it really be less appealing if it was plain Arial/Helvetica? Would you bet your money on it by A/B testing? I wouldn't.
These discussions end always in: design is important versus code is important. But that is completely not my point.
My point is that it is often, not very wise to spend disproportionately time to trivial considerations. I don't know your "proportions" so I may have no value judgment about that.
You're trying to refute the importance of the design through assertive statements like number 2 above, and yes, you are making the point that design is not important by calling the typography a "trivial consideration."
The fact that you don't understand why it's important to do typography well, and think that it's okay to just use Arial, doesn't mean that it is a trivial matter.
That's the kind of thinking that led to the look of Windows 3.1, and I would argue that while Windows might have been the winner at the cash register, it was a lesser product.
I applaud the Prismatic team for focusing on quality. MVPs have their place, but eventually you have to stop counting beans and think about making something great. That's hardly trivial.
I'm sorry, but I don't think you could be more wrong in this case.
Calling the attention to detail in design "trivial" is not just insulting to the designer and the company, it overlooks a very important fact—Prismatic is about the user's reading experience.
Two points:
First, just because they are able to explain the design rationale and historical context of their typographic decisions doesn't mean they necessarily agonized over them for months. A good designer knows these things from experience. I'm very happy that they took the time to explain them, and I don't think a blog post can be considered a frivolous use of a company's time.
Second, you seem to assume that there isn't anyone else in the company thinking about what you deem the "more important problems," or that they haven't already been through an earlier MVP stage where they focused primarily on technical issues.
Developers feel insulted and misunderstood when MBAs dismiss their efforts as fungible or underestimate their value, or the time required. It doesn't help when the equally vital role of designers is relegated to "triviality."
I would also like to disagree. In design, it's often the smallest details that make a particular design successful.
For example, Google Maps has a pixel or two of white around text. It's something you would never notice unless you looked for it, but it's part of the reason why gmaps have been so pleasant compared to something like MapQuest's maps.
Somewhat unrelated — I'm not sure why 'request an invite' bounces me to an OAuth page. Perhaps it needs to be clearer what's going on when I enter my email.
If you scroll down, you should see a bunch of info about how Prismatic works and why you are signing up w/ Twitter...and some testimonials from the community.
Really good call to omit Photoshop altogether from design endeavors like this. It's a waste of time copy/pasting actual data into your comp, whenever you want to change part of your design, you have to repeat it manually with all the other elements on the page, the fonts render poorly...
We lately try to go from sketch to simple HTML mockups. Having actual data readily available while designing is such an eye-opener. Thanks to web inspector, we can tweak the styles and see the changes get reflected in the entire page. Sometimes when we want to play with the layout, we take a screen grab, cut and move things around in Fireworks.
Nicholas Felton, about the Facebook timeline design: "You can come into a meeting with a very beautiful comp and it’s like, ‘Oh yes, we should do it that way,’. But you’re never going to know if you can do it that way until you pump in the real data and live with it for days or weeks".
I don't understand how Photoshop is still considered a web design tool. Unless designing a marketing site with no content, it hinders good design more than it facilitates.