Hacker News Reader

Comment by IIAOPSW | Hacker News Reader

Why do we care so much about design? Reddits success in spite of completely eschewing designers should prove design is over valued. So to does Craigslist. Japanese websites are proof by example that this is strictly a cultural phenomena. Why does Western society seem to value form over function so much?

How much is "good design" codeword for dumbing the site down as much as possible and making it shiny enough for people who get distracted by car keys? Reddit doesn't exactly have the learning curve of a command line tool or a rocket plane. Maybe I'm elitist, but if you can't understand reddit as it is maybe you're an idiot?

Call me an old man waving around my cane, but I say to hell with the redesign. Give me my plain old links in their plain old default blue underline and my mod-hacked-together css+bots. The internet need not anymore "progress for the sake of progress".

View on HN

> design is over valued

I would be more specific. Designing so things look good is over valued. Designing so things are simple to use the first time but don't restrict the user later on are under valued.

> Designing so things are simple to use the first time but don't restrict the user later on are under valued.

Which is exactly what I'd say they had and are now throwing away.

I wouldn't say these redesigns actually look very good... they're just easy. Maybe not to implement, but easy to imagine, easy to accept, and easy to make strict guidelines for.

And then you look other mediums of design, and you realize its only the software UI weenies who pull this out and think its something to be proud of. There's a reason you can slap bootstrap onto any website and have a "good" design; the result has nothing to do with the website itself.

and much harder to implement correctly

All the redesigns look Facebook-ish. Which make me wonder... Do the designers understand what people love about Reddit? The new interface looks void of information and way too distracting...

> All the redesigns look Facebook-ish.

This isn't by chance... Reddit is actively pushing themselves towards being the next Facebook, down to the whole "deprioritize subreddits/emphasize profile pages and groups of friends" mechanic. They just hope by boiling the frog slowly enough, people won't realize what's happening around them.

I strongly suspect they are aiming to sell sometime soon. The new rules, the new redesign, the attempt to facebook-ify the site.

Sell to whom? They already were bought in 2006 by Conde Nast

They want to IPO and cash out. They have explicitly stated that this is their goal.

Even this year's April Fool's event was called Circle of Trust and leans heavily on having groups of friends. Which, of course, sounds of so familiar.

Decentralized reddit using open protocols with each instance being a subreddit is going to be the next big step in decentralization.

It's about uniting forums without forums giving up their power.

Why do the instances need to be subs? It seems to me you just need NNTP extended to support voting and a few other features (most—probably all—of which could just be messages with new defined headers and, in some cases, no actual message body.)

I was thinking the other day. What if there was a browser extension that pulled in reddit/subreddit comment threads a la Alientube for the site or article you were viewing/searching. Literally anywhere you went on the internet for any hyperlink, you could check and see if there is a comment thread for that url. Decentralize the comment threads and related metadata onto something like IPFS and you have reddit but internet wide and you aren't limited to reddit's terrible search bar.

See this is why I don't do stuff like decentralized Reddit. What I proposed was my first idea but this is much better.

I wonder if decentralized voting for NNTP could actually be a good use case for some kind of block chain.

>How much is "good design" codeword for dumbing the site down as much as possible and making it shiny enough for people who get distracted by car keys?

You're confusing user interface design and graphic design, which are two different disciplines with different goals, neither of which is to be merely distracting.

>Japanese websites are proof by example that this is strictly a cultural phenomena.

You're assuming that Japanese don't consider that "good design." It is a cultural phenomenon, but your assertion that it's strictly a Western phenomenon is incorrect.

It's also particularly odd to invoke Japan as an example of a culture that doesn't care at all about design or aesthetics.

>Maybe I'm elitist, but if you can't understand reddit as it is maybe you're an idiot?

I don't think there was ever an overabundance of users who didn't understand Reddit, given the site's popularity. Like a lot of programmers, you're assuming an inverse correlation exists between the quality of a site's design and the intellect of its users. So yes, you're an elitist.

>Give me my plain old links in their plain old default blue underline and my mod-hacked-together css+bots.

... and here you are arguing in favor of minimalism, which is a form of design.

Your entire comment is just a rant about how other people's design preferences are less valid than your own.

"It's also particularly odd to invoke Japan as an example of a culture that doesn't care at all about design or aesthetics."

The belief that Japanese people all deeply care about such artistic things is largely a Western romanticization of Japanese culture, the sort that Roy Andrew Miller criticized in Japan's Modern Myth over three decades ago already. Yes, Japan is home to some interesting traditions, but that doesn't mean that broad society cares. And as Japanese websites show, Japanese mass society is quite comfortable with cluttered websites that one would be hard-pressed to call "elegant".

True, but I didn't claim that "Japanese people all deeply care about such artistic things", only that they had a history with a design aesthetic that differed in principle from Western design, but wasn't an example of the absence of design. One could just as correctly assert that most Westerners don't care deeply about design, either.

>And as Japanese websites show, Japanese mass society is quite comfortable with cluttered websites that one would be hard-pressed to call "elegant".

Do they look cluttered to Japanese people, though? Or do they only look cluttered to Westerners because Japanese text looks like noise rather than language? Admittedly, it doesn't look elegant, but that doesn't mean it isn't an example of design.

You're confusing user interface design and graphic design, which are two different disciplines with different goals, neither of which is to be merely distracting.

"Dumbing it down" = designing for user's mental models = ux design

>"Dumbing it down" = designing for user's mental models = ux design

Riiight... when Reddit designs their UI around the mental model of mainstream users, they're dumbing it down, but when Hacker News designs their UI around pg's mental model of "good hackers", they're being elegant and minimalist.

Most people will understand reddit if they give it a chance. Problem is that they're less likely to do so if they find the design and UX unintuitive.

I know I disliked reddit's design when I saw it the first time, and I've spoken to several people who never got into reddit because they disliked the design.

i would say that was the feature.

there was a similar discussion in emacs land, people were afraid they'd lose users, and to me .. that was perfectly fine. Those who stick to emacs in the spite of all the seemingly looking defects are users that really like it. Distorting something for market share like mentality is not good, it lowers entry but low entry means easy exit. Not worth it IMO. (plus in emacs, case, unless some large effort, no way a newcomer, even tech saavy, won't suffer, it's just too large)

Yes. There is a huge portion of people who will take one look at something that is mostly text, declare that it appears jumbled and/or confusing, and have any interest in doing anything with it immediately killed.

This large group of people, probably the majority, does not want to give such platforms a chance, and they do not want to get under the covers and figure it out. They simply do not want anything to do with something that is tightly-packed or primarily text-based. Many of the people I tried to recruit to reddit in the early days (either reddit.com or sites running the reddit software) had this reaction.

This can be a feature when you're trying to target specific types of people (cf. HN). But if you're trying to sell ads, it's a bug, because ad sales are all about getting as many human eyeballs on your platform as possible. Perceived ugliness is a real barrier to mass adoption.

I do agree mostly. To the point that reading reddit redesign makes me feel slightly bad.

Reddit value is so removed from the technicalities. Everybody even make jokes about its crippled search function. Yet nobody really cares.

And to an extent, the less designed a website the more willing I am to discover it. To me it means there's no love spent on css or trends, and much of it is spent communicating what the creator(s) wanted to communicate. Whereas most new websites have so much generic layers and presentation whooha, which is often just a revision of another website, I just can't care.

Haha yes users are idiots.

But as someone making mobile apps for over decade, "being elitist about it" is not going to accomplish anything. You'd think any rational mind would have no trouble figuring it out, and compared to what you're dealing with on a day to day basis as a developer, it should be a piece of cake. But people are stupid and are not magically going to wise up, and you're not going to change that.

UIs for the masses need to be as simple and foolproof as humanly possible, that's just the reality of it. Apple and everything they do is the best example of this.

It is a function problem, though. The reddit mobile "upgrade" is a huge regression in usability over just viewing the desktop site on my mobile devices, which I have done for over 6 years without complaint.

There is no upside to using this new mobile site. It loads more slowly, has obtrusive banner every time, has fewer features, is more confusing to use, and well, just sucks.

The problem I usually have with redesigns is the people advocating a redesign seldom pitch it armed with data showing the redesign will be better. Show me quantitatively that moving this button or changing that color or font size will move the needle on revenue, or engagement, or whatever the key metrics are.

At most software companies, if you want to add a feature, the burden is on you to estimate the value the feature will provide, and demonstrate how you will measure the value post release. How you draw the line from the feature to the value will be scrutinized. If the idea doesn’t stand up to it, it’s questionable why we need to invest in the feature.

This same scrutiny often doesn’t apply to redesigns. We are redesigning because it needs to look “fresh” or “modern”. It needs to “pop”. The old design looks old. So how do you measure “pop” in a way that’s not explainable by other product changes? (??) If it doesn’t quantitatively pop, do we go back to the old design? (the answer is always no) And the designers always get all bummed when you ask for this data. Look at all the criticism of Marissa Meyer having the gall to AB test 30 shades of blue. Why isn’t it ok to measure art?

I don’t think you’re an old man waving around your cane. This is an important question to ask because the effort blown on redesigns often could have otherwise been invested in things that would bring more value.

> Why does Western society seem to value form over function so much?

I don't think that's universally true, it's just a side-effect of marketing. Sometimes you redesign products to rekindle interest, but that rarely applies to other things (eg. public construction which is usually blunt but functional).

The trend I see lately is something I think of as dropoff panic. By staring at metrics too much PMs get worried about dropoff and start to focus their efforts on lubricating onboarding to such an extent in the hopes of preventing dropoff.

This results in the sort of dark patterns you see on Pintrest's signup and the overall dumbing down of everything into samey "friendly" interfaces that all feel like Facebook.

You manage to capture more of the casual users for the sort term by doing this, although they'll likely dropoff eventually. But this is done at the expense of your hardcore users who fell in love with your app/site/game despite the learning curve or dense interface.

Too much effort is placed on trying to capture a very low quality of user that likely wont engage anyway.

We do it to make products worse over time, to make room for new competitors. That'd explain a lot about the Web so far...

> Reddits success in spite of completely eschewing designers should prove design is over valued.

This is not a fair conclusion to draw. All that tells us is that design isn't the _only_ factor in success. There is nothing to suggest that Reddit would not be more popular were it better designed.

I'm not really sure what the cultural roots are, but the fact that visual appeal is valued by consumers makes it important to developers. I would also push back heavily on using Japanese web design as a counterexample. Their popular sites generally share an aesthetic, it's just a little different than what ours have. It's also worth mentioning that western designed sites dominate the Japanese market as well.

> Why do we care so much about design?

Because it's the lowest common denominator. Because everybody can have some input there. Because somebody in the org needed an "initiative" and the design is the easiest to explain.

We care so much about design because all the thousands of hours we've collectively wasted being confused by Reddit and Craigslist's interfaces could have been much better spent elsewhere.

Why improve anything? Because by doing so, we find what really should be.

That statement might hold if these 'modern' designs weren't oversimplifications presenting less information density and reduced feature-sets that heavy users depend on.

Here's the thing though - generally speaking, I don't know of many people who are confused by Craigslist. It's an effective design for what it needs to do - browsing is easy and obvious, searching is easy and obvious. It's just "ugly", which is a different thing.

I will say Reddit is a bit of a different beast - there's too many hidden and confusing features, and it's probably fair to say that iterating on the design was necessary.

My biggest concern is that what I've seen from the redesign seems to be focused on making Reddit less ugly than making it more obvious and functional.

We're both posting on what might be one of the "ugliest" discussion forums on the internet - but I've rarely been confused as to how to do something on HN, which I can't say for many, many "modern", "pretty" websites.

HN always felt like a minimalist reddit. Just a better tech subreddit

I like to use the parent/old person test for ui usability(obviously not every old person will do, you know the type tho). I showed my parents reddit and they picked it up right away. Both failed to learn snapchat and struggle through most "modern" websites.

I blew their minds when I showed them if they click the three lines on lpta of websites, really a hamburger menu, they get options and stuff.

Often users are just expected to know these "design paradigms" like symbols(hamburger menu) or which direction to swipe(im still guessing half the time Snapchat). Whatever happened to using words like settings to open the settings menu. When did GUIs' forget to show available commands to users in an understandable format.

Craigslist is one of the most frustrating user experiences I have ever had. Every attempt I have made to use it has ended with me looking elsewhere, and each time I have had vastly better experiences on other sites.

Everyone has a huge bias for what they already know. If you already understand and like Craigslist, I'm sure it's great. As a new user, it's awful.

Are you trolling?

This is interesting as I think the UX for these two sites is simple and honest and easy to use. Although in its limited functionality.

Can you give some examples of the hours you’ve spent being confused?

It may just be a different way of thinking for different people situation. For me, I prefer simple and honest vs pretty wit targeted ads trying to divert.

“Why do we care so much about design?”

Let’s say that you are at the grocery store and there are two cashier lines. One cashier is beautiful, one hideous. On which do you queue up?

I would expect that the answer depends on other factors- line length, speed of bagger, speed of scanner, type of candy bar in aisle, etc.

But those may be harder to measure and beauty is really easy.

There’s some research to show that attractive people make more than unattractive when controlling for other factors- https://www.ssc.wisc.edu/~scholz/Research/Beauty.pdf

I think this is part of a bias being transferred to products. Pretty people are favored more, so pretty web sites are desired. Although data can show this doesn’t matter (Craigslist, old-good-Reddit vs new-shitty-Reddit) very much.

> One cashier is beautiful, one hideous. On which do you queue up?

I'm going to pick the one that scans items faster, accurately displays receipt information on a screen before paying, accepts chip, automatically prints relevant coupons, automatically adds product recalls on the receipt, and doesn't nag me to sign up for a custom credit card.

> accurately displays receipt information on a screen before paying, accepts chip, automatically prints relevant coupons, automatically adds product recalls on the receipt, and doesn't nag me to sign up for a custom credit card.

None of these (maybe save for the last if you have a cashier that isn't doing their job) is going to differ between the two cashiers as they'd be store-level policies.

Sure, but the metaphor is about pretty vs. shitty UI, not pretty vs ugly (all usability criteria equal).

I agree that those are more important factors. How do you measure that when you’re choosing a line? How much time do you spend measuring before it’s not productive.

> Why do we care so much about design?

Well, in the article they specifically say the site is confusing to newcomers, and they're trying to fix that problem. So that's why.

It's what the owners do when they don't want to confront the real problems. In Reddit's case, their social engineering system has failed, and they don't want to admit it. YT redesign in 3.2.1...

Youtube already redesigned last year: https://www.theverge.com/2017/8/29/16216868/new-youtube-logo...

Reply on news.ycombinator.com