Why “Simple” Websites Are Scientifically Superior | Hacker News Reader

Why “Simple” Websites Are Scientifically Superior

soshitech.com

80 pointsBrajeshwar29 comments

So let's talk about Facebook. Over the years the site and service has become bloated beyond belief.

I think part of the reason people (especially younger audiences) are using Instagram/Twitter/Snapchat for key functionality over what FB provides is because of the simple UI/UXs. Kids might be seeing FB similar to how they see Excel spreadsheets. Complicated shit for adults that isn't synonymous with "fun".

No; there are a lot of problems with Facebook, but I'm pretty sure it boils down to parents being on there. That, more than anything else, will keep kids away.

You could also argue that Facebook focuses too much on content consumption whereas Twitter, Instagram, and Snapchat all focus much more on content creation.

Another problem I've seen a lot lately is the Newsfeed algorithm just being downright terrible. I've noticed a rash of comments on months-old photos lately where people are saying "how did I miss this photo?!" If you're going out of your way to create content and no one sees it, what's the point?

Agreed, and I think it's mainly because it gets hard to figure out what content to display when 1) people are following 300 friends + 300 pages; 2) advertisers are paying for their content to be seen.

The competition for a spot in the users' news feed makes it increasingly harder to present the user with quality content...I sometimes wonder how it's even possible to test the quality of their algorithms a such a large scale (kudos, btw), even with all that automated continuous learning.

The article states that once a site layout follows the archetype for that type of site then the brain is "happier". I would say that at this point Facebook is so ingrained in many(not all) users that its design follows a known archetype(The FB design).

Facebook has also simplified their design considerably over the years.

Do you remember when you had all those wall plugins like drawing apps, and "SUPER POKE" it got way out of hand.

And part of the reason Facebook did so well when it first came out, was because it was a far easier alternative than the CSS bloat that was on Myspace.

Remember going to your friends page, and three videos would autoload, with 7 songs playing and glitter graphics?

Nothing from "Why Simple is Scientifically Easier To Process" down is backed by any of the actual research papers he mentions. He's also mixing studies of (at least) two different things here. The Google and Harvard studies both looked at flashing up images of websites, and were about judging aesthetic appeal, not recall. Meanwhile, the papers on fluency are about recall of information presented (items on a restaurant menu, for example). The former set of papers don't mention any connection with Miller's 7+/-2. In fact, those papers don't posit a physical basis: they're empirical studies. This stuff about photoreceptors seems entirely irrelevant.

If you want to do something scientific, why not actually test your website? It's what the researchers in those papers did, but it's missing from his recommendations.

Maybe it's because SoshiTech is an inbound marketing/content marketing/content farm/black hat SEO website? I clicked on the "Apple" tag in the tag cloud and got back a list of completely unrelated posts. TBH I don't understand how posts like this get so far up the front page without some shenanigans going on.

Yes, this content was scrapped. (Hi, I published the article originally)

While I did re-distribute the content to Medium, we very much practice these principles where the post was originally published on ConversionXL

While the studies referenced do pull from different areas, "beauty", "retention", "conversions" they are all used in order to back up that science supports simple, prototypical, sites performing better.

This means less visual clutter, everything being where it is "supposed to be" and presented in a way that isn't over stimulating to the senses.

As far as the photoreceptors in your eyes are concerned, that's the bridge between the real world and your brain.

You don't actually "see" anything, your brain just decodes the information.

Less information to decode (i.e visual complexity) the easier the actual message has to get to your brain.

>Even though this was part of a bigger growth strategy, the results are still impressive, over a million new users have been added since June, when the new logo was first debuted.

No, the results are meaningless because they're confounded by a big whopping set of hidden variables.

This is like extolling the virtues of vitamin C by pointing to a cancer patient that was put on chemo and also given supplements to keep them healthy, and saying "Even though it was part of a bigger treatment plan, the results are still impressive; the cancer has been in full remission since June, when the doctor first started administering vitamin C."

I prefer simple HTML pages over JavaScripty type pages. And it seems these js UIs keep replacing perfectly good HTML UIs. The Hadoop web admin UI did this going from 1 to 2. I think the new google groups did the same.

Agreed. I know of a forum that has barely changed its design since 1997-1999. It loads fast and has many members. The focus is on content over appearance.

These javascript-heavy UIs feel heavy, almost syrupy, by comparison.

Agreed. I like the simplicity of HN (though the expired link thing is quite annoying.

> I prefer simple HTML pages over JavaScripty type pages

How can you compare html and javascript? It's like comparing apples to oranges!

Science can't tell us whether a design is superior. It can tell us that the design performed better in some specific test (such as a beauty rating), but leaping from that to blanket superiority is not science.

Yes! Thank you, I get annoyed when people think there is one definition of good design/art/beauty ...it's a thinly veiled display of narcissism. It's like trying to tell someone what kind of food they like, or who they find attractive- there are trends, and sometimes it seems like there are rules, but nothing is ever 100% universal as far as I can tell.

The article is discussing "simplicity" as one aspect of design. While I agree different designs can really only be viewed subjectively there are some aspects of simplicity that can be judged objectively.

It's pretty clear that the superiority metric is going on what the site's intention is: conversion.

So while you're right it's not superior in every metric, it's superior in the only metric that the people creating the sites care about.

Scientifically superior is a strong claim.

Though what is important to note is that web accessibility is still important for visually healthy person. It sucks that visually disabled people have trouble navigating the web, but heck as a sighted user I have trouble finding things too. Just yesterday I said how awesome the original Mac Writer was. It was simple. Compare to today, I can't even locate the line space option easily.

I can careless about the pretty round corners. I care about efficiency. Is the information adequate or not, is it loading quickly or not.

Not all simple websites are created equal either. I have been to a lot of static blogs lately and friends, half of them have horrible nav or CSS choice.

Web semantic is scientifically superior because then a program can parse the input very easily. No more div soup to create buttons. img should have alt. <p> shouldn't be misused.

One wonders if the HTML works for our purpose or not. It was meant for document sharing. Does it work for our multi-media purpose?

And yes, I can't do all the things above because I just do what works... if it works, it's good right?

Having a simple site could also be backed by psychology studies. Hick's law states the more choices available, the more time it takes to make a decision. To test the theory, two jam stalls were set up in a market. One with 50 choices and the other with 3. We can all guess who sold more.

I really love the direction UI/UX is going in, more and more based on science. There is already good amount of work done in the human factors and ergonomics area -- I would love to see this knowledge applied to web and mobile apps.

This is not a new direction. UI design was scientifically-based long before the phrase UX was coined.

Just a couple examples: -The CHI conference started in 1982, and the content has always been primarily scientific research on UI -This book from 1983 has been very influential on the evolution of UI: http://www.amazon.com/Psychology-Human-Computer-Interaction-...

The field of Human-Computer Interaction has been around for decades but the larger design community has not always paid much attention to the literature or embraced the scientific approach to design.

So I've seen this article in 3 places today.

* ConversionXL * Medium * and now Soshitech

You mean, like this? http://ludens.cl/paradise/turbine/turbine.html

That was an extremely non-simple webpage to say that.

Content was scrapped.

TL;DR -> DMMT

(don't make me think ;)

What does "DMMT" stand for? You made me think, dammit!

Is this your summary of the article?

Reply on news.ycombinator.com