There's a kind of fetish in design for modernism, futurism, and this kind of "sleek commercial humanism" (someone coin a better phrase).
The cliché is a self-taught designer in a minimalist loft in some gentrified inner city, drinking espresso from an expensive vintage machine, obsessively cleaning their wooden slab desk, listening to post-techno or whatever...
...making designs that look like they signify design... with some new or rediscovered modernist sans serif font... matching pastel colors... tasteful transitions... high-resolution photographs that signify urban creative prosumerism...
It's a kind of functional tradition, but as a whiny outsider I'm really getting sick of it. So I like this article because it's just highlighting that there are other values to aspire for, not just these matte white ideals of OCD modernism.
Reminds me of a video from some old RubyConf, I think, after a presentation about how "beauty corresponds to quality", some humanist in the audience asked a thoughtful question like "the type of beauty you've explained seems mostly like a classical or modernist ideal; what about stuff like postmodern notions of beauty, could that be relevant too?"
(Ruby itself is ugly and messy, a weird mix of Perl and Smalltalk... Hacker News is ugly and messy... even the interfaces polished to look simple and clean turn out to be ugly and messy when you look closer, so maybe part of it is just about honesty.)
The minimalism in the examples is a brand statement and a form of social signalling.
It's UX - in the marketing sense - rather than UI. And yes, a lot of web marketing assumes the buyer is exactly the kind of person you're suggesting - maybe because often they are.
Craigslist, Plenty of Fish, and all the other "Can't be bothered to add styling" sites are taking the opposite approach.
The lack of overt social signalling becomes an inclusive social signal of its own. By making the design as functional as possible and not aiming for any demographic, they're more likely to create something that's acceptably styled and useful for all of them.
Trained designers don't seem to understand this, because although they pay lip service to the idea of usability, they're mostly only interested in designs that include aesthetic signalling.
In fact there's a whole tradition of professional 'get shit done" design in hardware and software which gives users all the options they need at the same time with minimal fuss. This kind of flat design can be very productive, but it rarely looks pretty.
Tangentially, I think FB has the worst of all worlds. The aesthetic choices are poor. The ergonomics are poor. Critical settings are hidden. Content appears more or less randomly. You have limited control over what you get.
FB is only popular because of network effects. If the design was less dictatorial, it would be more popular still.
Trained designers don't seem to understand this, because although they pay lip service to the idea of usability, they're mostly only interested in designs that include aesthetic signalling.
This is a very common human pattern: Confusing signalling+lip-service around a complicated and subtle issue with genuine understanding. I suspect it's a common pattern because it's a part of the learning process.
Ruby might be messy, but I disagree entirely that it's ugly. It's one of the most enjoyable and human programming languages I've ever used, and is what I work in daily with much enjoyment and satisfaction.
I was in love with Ruby since way before Rails, and I still see the sparkle in its syntax. It's a strange kind of beauty, but beauty is weird. A horse can be ugly and beautiful at the same time...
As in, my Ruby code generally has the least amount of implementation cruft compared to other languages. It would seem that it fits closest to how I think.
Hashtables, resource locking, regular expressions, string formatting, are all things I have to deal with a lot and I like dealing with them in Ruby more than any other language. Also, reflection in Ruby is superb.
The old https://rubygems.org used to include the steps to build a gem, and publish it[0], which was helpful.
But in their redesign they removed it, so now that info is hidden behind two clicks (Guides -> Publishing your gem), but that page isn't near as concise, and doesn't show you how to build the package.
I've found it to be more of a cycle. The current trend in systems and interfaces is minimal with people taking it too far. It will swing back to more comprehensive interfaces (ui, api, etc) when people get tired of not having anything and comprehensive will be back. The downside is that comprehensive interfaces will be taken too far (eg: the 1 api call to rule them all) and then people will introduce minimal again.
Good system/interface design is rare, and good design works well regardless of the current trend.
Do you also have this feeling when you use some web page, and the someone decides to do UI refurbish, that many times, it will look better but the feeling when you actually use it become worse? I'm not talking about simply becoming used to the old, and resisting new. It's rather that many times in the first version of the UI, many things become clear once you see it, but making it more aesthetic means many times to hide or obscure some otherwise natural or common actions.
Maybe it's me, but I consider Craigslist's interface to be clean and minimal - with no unnecessary stuff (like stock photos, or animations, in the background). Aesthetic value is not always the priority, from the user's perspective (heck, I want to rent a room, not - visit an art gallery). Looking great at the first glance of a non-user is a totally different thing from a site serving its purpose well for its users.
And exactly for the pursue of the minimalism I went from Delicious to Pinboard. Cleaner... (OK, plus - working).
Also, it's why I prefer HN looks to Reddit looks. Or even aesthetic looks of things like Blizzard forum.
And visual design is all about understanding how layouts, colors, typography etc contribute to a larger sense of what the thing is about, its values, history, community, etc.
If Couchsurfing looked like Airbnb, that would be a design failure, probably. Or at least it would alienate many of its existing users.
Couchsurfers typically don't mind a reasonably messy room, because it's all about thrifty hospitality and this student/traveller community.
Looking at Couchsurfing now though, in fact it does look like Airbnb... Like they "modernized" the design... and indeed I've heard from former big fans of the site that they don't like it anymore, that it's changing culturally, and that they now prefer BeWelcome... and their site doesn't look like Airbnb at all.
With a lot of SaaS landing pages I just wonder who they imagine as their target market.
Maciej at Pinboard seems to be doing good and earning respect by designing the site he and his friends would actually like to use.
Craiglist is already an established player, and its minimalism is part of a well-known brand.
Suppose you were to pitch a new startup today, at YCombinator or Product Hunt or a VC meeting, and your web app copies the Craigslist aesthetic. Would you be taken seriously or laughed out of the room, even if the functionality and content was top-notch? Would you take that chance?
Because that gives you some degree of design. There's an enormous difference between "looks the same" and "looks like 1996". The average end-user has basic expectations and will very likely leave if those expectations aren't met, at least until they've already bought in.
I am more likely to leave if the site looks overdesigned, honestly, because that is a signal that this company is focused more on the "acquiring customers and extracting attention" side of the business than on the "providing a useful service" part I'm actually looking for. If the site looks plain and practical, it might turn out to be useful; if the site looks gorgeous and cutting-edge, the company is probably just going to piss me off.
If the site is plain, it looks hand-crafted; like somebody who cares about it made it, and they're too busy making awesome content to fuss about the paint job.
If the site is "designed", it looks like someone just hired a designer to sell me something.
For certain kinds of site this look could be a positive - thinking of Pirate Bay, 4chan etc. Not the sort of things that get VC funding granted, but where a non-corporate, amateurish design gains you street cred.
At the airshow, no one ever looks at the F-16 cockpit and says "that is not a beautiful||fresh||clean||simple||minimal interface". It is understood that the interface is for use by a highly trained user in an environment where usability is very important.
Highly usable environments require some sort of training. The amount of self training that someone will do for something like a web site will entirely depend on how motivated the user is to achieve the definite result that is the ultimate goal.
The question then is; do pretty websites cause motivation for further engagement? I suspect the answer is yes, at least a little, at least for now. In the future that probably won't be true, people will get tired of anything anyone can possibly do in terms of appearance and will value usability more.
Highly usable environments require some sort of training. The amount of self training that someone will do for something like a web site will entirely depend on how motivated the user is to achieve the definite result that is the ultimate goal.
Is it possible to ramp complexity while ramping motivation? Games seem to do this quite well.
Call it "pretty" to signify the idea that the only things designers care about is aesthetics. But I think designers care about building an emotional connection to the customer and ensuring that only the right functionality appears in a given context. I feel like using the word "pretty" is code word for demeaning the value and thought that people put into creating an app/website beyond the functionality. Oh it's only a thin veneer who cares.
All too often, people want to cater everything to everyone so they just throw all the options on the page and expect you to know where to find what. Or they don't think you're a human and should just be happy with whatever you're given because hey, the functionality is there.
But as consumers get used to a level of design through the apps/websites they use on a day-to-day basis, the ones who do care are going to want that sense of quality in their other tools.
I don't love the F16 analogy. Some things do require highly specialized training. The vast majority of websites and applications don't. Furthermore, there's an additional layer of difference since physical buttons/gauges have an advantage of easy access as compared to layering functionality in digital interfaces.
I enjoyed this article, and overall agree - sometimes you want / need a mess.
Craigslist for example is not unbeatable, nor is it a messy interface I think is truly good. Really, I think it's more a matter of who they are at this point, and that they have a user base that gets things as are, and they can't change now. New users will deal with the mess, because they more or less "have" to use Craiglist for something. Not that craigslist needs to be pretty, but being frustrating for even regular users is not a good sign.
Photoshop is a better example, I think. Its ugly in the sense that there is a lot, but whats there IS actually designed; well mostly.
Facebook is trickier. They have a network effect which drives many new users to overcome on boarding, but also people view it on mobile a lot, and so don't see that interface. And those that do, probably have blinders and only see the second column.
Really, I think there is a time, place, product and audience for it. It's important to think about what information you "bubble up" in an interface, and what you require them to click-thru for. Obviously if users learn your messy UI, they're going to be upset with complete redesigns, so don't do that. Do care about the look of the interface, it always matters, but sometimes "matters" is pretty, and sometimes "matters" is functional and complete. If you can strike a balance between the two, and create an experience that your users can understand and enjoy, you're on the right track.
IMHO, too many designers confuse having lots of controls with "messy interface". If the controls are poorly laid out then it certainly can be messy, but if they are organized and arranged properly it is not messy. You don't say a library is messy because they have a lot of books, those books are neatly organized and indexed.
People have gone way overboard in hiding functionality in modern design. Discoverability seems like a dirty word these days. The ironic thing is that by removing the obvious controls for so many of the features in the name of simplifying the interface for new users, you actually make the interface more complex and require more training on the user's part to operate it.
The days of browsing through menus looking for an option that vaguely sounds like what you want seem to be gone. Now you have to just know the magical screen gesture to make it do what you want. The only way to discover it is to know that such a feature already exists and Google for it.
Hey iPhone users, did you know you can swipe up on the lockscreen to bring up a menu full of useful controls? I'd say roughly 1 in 2 iPhone users I've talked to know about that feature. Do you know how you were supposed to discover it? Because I don't.
> You don't say a library is messy because they have a lot of books, those books are neatly organized and indexed.
Well said! I think the same thing about magazines and newspapers. Although we should not be so quick to compare; part of google's success was their simplicity, compared to yahoo's complexity.
> The days of browsing through menus looking for an option that vaguely sounds like what you want seem to be gone.
Maybe. I'm doing something like this within my app, but for anywhere complicated I just have a dropdown with a bunch of examples that they can start out with and tinker with from there. I think that can go a long way in discoverability and understanding how to do something, and whats possible.
Photoshop is a thing for 'doing'. The crowded UI disappears when you actually use it, things flow very nicely. Other graphical editors on the other hand ...
As someone who's never used Photoshop, I don't understand the GIMP hate. I use it all the time for a variety of imaging tasks, and it rarely surprises me - everything's discoverable, you don't need to memorise weird shortcuts, and the single-window mode in 2.8 removes the biggest pain point people were complaining about. The interface is also extensively configurable, with pretty much anything dockable into anything else (not that I use this feature much).
Doubtlessly many GIMP haters, myself included, are frustrated because GIMP made many UI decisions that are incompatible with Photoshop without offering any compelling improvement. For example, zoom in PS is ctrl-minus and ctrl-equals, whereas in GIMP it is minus (no ctrl) and plus (aka shift-equals). This change is very annoying because zooming is one of the most common actions in photo editing and becomes part of your muscle memory. This design choice, and many others in keyboard shortcuts, menu layouts, etc. contribute to a feeling that GIMP tried to break compatibility with Photoshop on purpose for no reason. It feels like GIMP is actively hostile to Photoshop users. It also feels like the GIMP developers were very arrogant to think that following their design whims was more worthwhile than being compatible with a well-loved industry standard program.
For serious photo editing, GIMP is completely useless because it lacks the Adjustment Layers feature that is critical for non-destructive editing, and it did not even support 16-bit color until a few months ago.
GIMP is one of the huge failures in open-source software IMO. They should have just made a Photoshop clone.
Nothing makes sense for me in GIMP. Don't call me grumpy too fast, I used almost every computer graphics related programs and none gave me this level of frustration. Be it 80s dos painting programs, x11 image viewers, windows node based compositing, 3d animation packages, even illustrator (which has some odd ui abstractions sometimes).
I thought blender changed dramatically in term of UI / ergonomy. I used to say that blender is the GIMP of 3D. But I came with a strong Maya/Houdini superiority complex.
GIMP needs to stop making itself unique from Photoshop and try to emulate it as much as possible. There are way more photoshop users in the world than fresh users going into GIMP.
Neither Photoshop nor Craigslist are messy. Everything is very well organized, framed, boxed and so on with a logical hierarchy (which in PS's case can be collapsed almost entirely to display only a simple interface). It's the very opposite of messy.
In other words, you want the simplest interface that allows you to do what you need to do.
> In other words, you want the simplest interface that allows you to do what you need to do.
From which follows the thing I've been repeating over and over and over again here - all this simplification of interface comes at the cost of cutting down features to the point the product is barely usable for anything at all.
I actually remember some heated discussion on changes of the UI in Photoshop not that long ago on this forum. So even there (re)design is an issue, and lovers of the 'little=beautiful' design have their influence.
Blender had an interface that was, to some people, very messy, but other very experienced, knowledgeable users fought tooth and claw to keep it - and ultimately lost. The changes were performed, and, after a while, ~everyone was happy.
These kinds of interfaces get some effort to get used to, and afterwards, you learn to love their quirks and don't want them to change. New ones get added / moved / removed organically through the life of the software, so the user and the UI change together. And all of that creates an emotional attachement.
For craigslist specifically, I think people often confuse "simple" with "pretty". Craigslist is a remarkably simple interface. I've never seen anyone - regardless of age or proficiency with computers - get confused while using it.
It does what it sets out to do, without pretension or fluff. If that's considered unsimplistic, I don't know what the word means.
I think the argument is a little different for Photoshop, but I'm not a user and frankly I do find it confusing. But like the mixer analogy, a professional tool should sometimes sacrifice simplicity for utility.
> I think people often confuse "simple" with "pretty".
Oh man, this.
Even richer is confusing usable with pretty. I guess nowadays people say "user experience" instead of usability, or the extremely unfortunate: UX.
Simple usually covers the very first prototype version, but then there are always requests for complicating the feature set so you can't have a simple interface anymore.
Joel Spolsky had a conference talk about this once (can't remember where I saw it). There is a tension between simplicity and power: some need simplicity; others, power. It's a spectrum. What developers and designers should strive for is elegance so that one can move along the spectrum without noticing it.
I just love those kinds of interface. I don't think I would enjoy using them, but I love the history that sometimes go with them. Typically they're in-house apps that just get extended until they do every thing, and only an old guy named Gary knows how everything works... sort of.
Crazy as those UIs look, there's a certain method to the madness. Colors and varying shapes help create spatial memory which helps with data entry.
The person who works with this every day knows intuitively that "pink box, bottom row, second from left" means something specific without having to read the labels.
If this UI were redesigned in the prevalent iOS 7 style with uniform design elements, abundant whitespace, very little contrast and lots of text labels in a light typeface, I think it might actually reduce usability.
I think GIMP has a great user interface, it enables me to work fast, do things quickly, with high precision. And I am one of those who switched from Photoshop to Gimp. It is not because GIMP is better, but it has all the important features I need and it runs on Linux.
You are conflating things. You switched to Gimp because it runs on Linux and is functional enough to replace Photoshop, not because it has a great user interface. If there was Photoshop on Linux, you'd be still using that. Functional ≠ great user interface.
I think that simplicity still applies, even in interfaces which aren't considered simple.
It's just a matter of how you conceive simplicity. John Maeda talks about "thoughful reduction". It's not less for the sake of less; you take away while it makes sense to take away. That's simplicity. Clutter isn't the opposite of simplicity. Simplicity has to be conceived always factoring in the scope of the design. If something is necessary and you remove it, you're not making your design simpler, just dumber, limited or crippled.
tl;dr: simplicity will look different in different apps. Photoshop will always look more cluttered than the prospective and portfolio-enriching weather app which is featured on Dribbble.
In my opinion, what's important to some categories of users is high information density. This is what makes the 'ugly' Craigslist design useful to its users, the ability to have a bird's-eye view of potential navigation paths throughout the site because it reduces the number of steps needed to perform towards a certain goal and makes evaluation easier, without distractions.
There's a difference between interfaces that are messy and those that have high information density.
Also, there are quite a few sites which benefit (profit) from throwing off the user's train of thought and injecting unrequested information:
- Go on Amazon looking for a product and you'll see lots of other products that may be relevant but that are totally not what you were there for.
- Facebook being an advertizing company is of course happy to inject promoted posts to satisfy their clients.
Edward Tufte has written at length[1] about information density, and has shown that it is possible to design interfaces that contain a huge amount of information AND yet are clean to the point of being beautiful.
Pointing at popular products with crappy Ux and implying that their popularity vindicates their interfaces is totally specious. Does the author honestly believe that that is how markets work?
Those products all have years of growth and hefty network effects backing them up. You can infer exactly nothing about the importance of ux based on the difficulty of unseating them.
The reason why these complex products win isn't because of their design, but instead because they have moats around their business. Craigslist has network effects and photoshop is an industry standard product. The author is confusing economics with design. The underlying economics are what make these products durable, not their complex design.
There is nothing messy or ugly about the Craigslist page.
It is unadorned. Unadorned isn't a synonym for ugly.
It is neatly organized into a clear structure which stays consistent and in which you can easily find everything. Therefore it cannot be called messy.
Yes, people love presentations which are neatly organized into a clear structure. So if you falsely equate that with ugly and messy, then you can argue that people love ugly and messy.
Suitable word definitions of words make almost any point come true. If not, you can always bring out the bigger guns, like outright equivocation: shifting the meanings of words while midway through making the point.
This sort of recalls some nostalgia i have for my middle school days. this was pre-Facebook, around the myspace era. And the prevailing thing on the web were these baroque, meticulously designed websites composed of handcrafted photoshop work, several kids in my class had designed and built their own websites and they were incredibly unique, detailed, and interesting to look at. Idk if anyone remembers this site but there was a site called "spoono.com" that was basically just an aggregator of these meticulously handmade, pixel perfect websites.
The most striking thing for me is that during this period the web felt much more like physical space. I would go to some of these websites just to spend some time immersed in the atmosphere they had so effectively created with this old, pixel by pixel style of design.
The biggest effect of html5/css3/social network era for me is that i no longer give a shit about your website. I might end up on your website if i click a HN link, but the desire to be on your website, with all the sense of place that entails, is completely gone.
And all of these websites are completely gone now too, vanished without a trace. It's like an entire generation of the webs history has been erased.
I'm glad I'm not the only one that realized this. Not that I actively go around trying to find other people though.
Even bad geocities pages were unique and interesting. I haven't come across anything that's like that in so long. It seems like an endless cycle of content aggregators and the usual news website. There's really no reason to actually explore or any easy way to do it. Might just be me though.
Even myspace allowed you to edit the CSS and pretty much do whatever you wanted with your personal page. Now all we have is Facebook and the actual information isn't important at all. I rarely even bother visiting people's pages nor have I updated mine in so long. It just doesn't matter for the most part.
The reasons why Facebook doesn't improve its horrible interface have to do with incredible technical bloat, and their userbase being stringently resistant to change. It's the type of person who won't start using something else until what they have doesn't work at all anymore. I predict an exodus if they tried to modernise.
Even so much as flattening out some of those beveled edges and gradients would be enough to throw people off.
"Ugly and messy" doesn't have to be a part of the equation when what you want is "dense." It's possible to have minimal, clean, dense design. With lots of buttons and features, beneath a simple UI.
If the utility of the application provides more value than the pain of the interface then people will use it because it's worthwhile; in other words users will put up with the crap if they're getting enough value out of it.
It's worth noting that all of these examples are world-class, industry-leading applications in their respective domains. People use them in spite of the interface. If your app hasn't proven itself to be in that class then an ugly, messy interface isn't going to work.
> If the utility of the application provides more value than the pain of the interface then people will use it because it's worthwhile; in other words users will put up with the crap if they're getting enough value out of it.
True in theory, but I don't think this is the case with Photoshop and Craiglist. Maybe with phpBB though. But in the two mentioned in the article, any serious attempt on making the interface prettier would most likely kill the usability. Those products are tools. The interface is there to let you get the job done quickly and efficiently.
And yes, you may have to spend few minutes learning how to use it. Just like you have to learn how to use a microwave, or a car.
The current trend of "beautiful and simple" is driven by the market - beause companies don't care if the product is useful for anything else than getting you to sign up (and sometimes start paying - startups that aim for an exit don't even care about that). You can clearly see how little a company cares about solving their customers' problems by looking at how much they prefer beauty and simplicity to functionality.
Agree with this completely. Hacker News, Craigslist, Reddit all wildly popular because they are functional...not pretty. They get you the information you need and the actions that you can accomplish quickly with no fuss.
Minimalist makes complete sense on mobile and I'd wager that most of what's driving the "minimalist everything" is responsive layouts.
I agree with the sentiment of the article that aesthetics are often mixed up with good overall design, whereas good functional UX design is an even more important part of it IMHO (of which craiglist is an excellent example).
I don't like his usage of the word cluttered though, which implies disorder and bad UX design for me. An audio mixer or Craighlist are not disordered, but have quite a lot of order. I guess he means dense interfaces, which can be the right thing to do for a UX design, if use cases demands it.
A related observation was made by Northcote Parkinson in his famous book "Parkinson's Law" (not sure which one, there seem to be different versions). He noted that institutions that have huge clean spaces are essentially dead, as opposed to institutions which are organic and untidy, which was a sign of vigorous activity. Maybe that's somewhat true for websites and applications too..
And I would also point out to Emacs and Vim, which have visually terrible interfaces, yet some people swear to them.
Honestly I think this guy misses the point. A lot of the extreme minimalism we see today is more about visual aesthetics rather than a clean designed meant to make an application easy to use.
A messy interface is NOT a good thing. What the author seems to try to imply is that minimal interfaces means you lack of functionality or capability. That is a misconception. It is perfectly possible to make complex interfaces with few features or simple interfaces with lots of features.
You can e.g. simplify a interface by making it easy to combine functionality and you can complicate it by making lots of overlapping functionality which can't combine.
Another way to make something complex without adding features is to always display all possible actions even when they are not relevant for the task you are presently doing. Clean interfaces strive towards only showing the actions relevant to what you are presently doing.
Simplicity is overrated. Experts often need more than the simple mantra 'do one thing well'. What is often needed for the user (novices and experts) is clarity. An unambiguous, clear interface is much better than a simple one. Even a simple interface can be confusing.
The compliments.dk site used as the opening example might be "beautiful", but the first thing that struck me was the unreadable white text on partially-white background (e.g. the "products" and "menu" text). That's looks like a case of misplaced design priorities to me, and is simple enough to "fix" e.g. with a faint drop-shadow around the text (I say "fix", since it wouldn't help user agents which don't implement the relevant CSS)
Now, obviously I’m not suggesting you should go clutter up your design work, or make it look crappy on purpose. I’m also not suggesting that the examples above couldn’t be improved.
My point is: there is no single right way to do things. There’s no reason to assume that having a lot of links or text on a page, or a dense UI, or a sparse aesthetic is fundamentally bad — those might be fine choices for the problem at hand. Especially if it’s a big, hairy problem.
Interestingly, I thought that was exactly the case at first. Since the domain of the link is actually signalvnoise.com, Privacy Badger blocks most of the requests to medium.com and leaves the article as unstyled text. I would have been perfectly fine with this, but the images got really confused for some reason and would only load as tiny ~30x30 thumbnails.
I'm not sure whether to blame Privacy Badger or Medium, but this isn't the first time I've had problems with the funky progressive scaling thing Medium does with images.
CraigsList and Facebook are neither ugly nor messy. They're functional given their needs.
Maybe the thesis of this article should be:
Don't let your designer go berserk and just make whatever white-space-heavy-drop-shadow-gigantic-photographs-four-words-per-screen mess they feel like if that's not actually what your users need to find satisfaction in your site or app.
Seem that the author of the article is comparing information dense interfaces with those that aren't. Not really a fair comparison.
Of course they look 'messier', in as much as there's more on the screen. People use spreadsheets --yes, in 2016!--because they're the best way to see some types of data, even if the interface is horrid and outmoded.
I like how you show an mixing board for your example because that industry is actually getting a small upgrade.
Least I heard Yamaha hired a bunch of people to redo the interface on some of their higher end equipment. Trust me, they need it. When you need to go to 3 different menus to configure one setting properly then there is a problem.
Well no, there are some features that were unthinkable to do in such a small table top setup as they are in the digital setups.
Some of these are virtual EQ racks, piping audio, and doing some other cool stuff like eco reduction and speaker sound offsets based on distance from crowd.
Here's why I like them: they're like Odysseus's sirens, but for hipster UI designers. SV is littered with the wrecked hulls of startups that were like, "we'll just do Craigslist, but with a hip, clean, modern UX. It'll be like taking candy from a baby!"
As for messy, that comes down to it getting in the way of getting something done.
If something has tools and parts all over, but the workman knows exactly where everything is and can reach for them as needed, it may look like a mess to an outsider, but will be efficient in operation to the workman.
I see way too many designs that mistake clean for functional. Apple is a huge offender in this category lately. They have a terrible habit of hiding functionality with no hint of how to interact with it.
Content-wise there is no competition for the examples he came up with. I think their success should be credited to habit (people are now used to it) and content, not to their 'ugly' interfaces. I think that new apps / sites like Uber and Airbnb wouldn't have succeeded with Craigslist's aesthetic and that the iPhone would have flopped if the interface looked like Windows mobile.
And for a personal example, I come to Hacker news for the content and the community. I do not like the look of it.
I'm thinking that two smaller images shown side by side with captions underneath each would have made the point more crisply. I, too, missed the carried over point while scrolling down the first time.
And if you stick to that design principle, then any non-trivial software will quickly end up looking like Photoshop. And it's good. This principle is how you make tools. They are there to support your work.
Do people use bootstrap for anything more than a landing page? I see what the author is trying to say - but the kind of design that they're highlighting as the "new simple. beautiful" interfaces is rarely used for any thing but something nice to look at before you get into the meat of a product.
I think the article is still spot on - the "minimal design" frenzy has influenced more than just bootstrap landing pages.
The YouTube App on my iPad looks wonderful, but for the life of me, I can't find any back/forward navigation buttons. I suspect they were minimized away (I'm on iOS7 on my iPad2, this may be different in newer versions).
My Android phone (Cyanogenmod) has a wonderfully minimal interface that even includes a universal back button - but a "go forward" button was presumably deemed too messy and either thrown out completely, or hidden so deeply that I did not manage to find it in a year. This is especially bothersome in the Android browser...
Yesterday I used a personal Oxygen sensor, for safety reasons. It's a wonderfully designed piece, with just two buttons. Yet their functions are so overloaded, there is no way you can use the sensor without studying and memorizing the user manual.
My Windows PC becomes "simpler and more beautiful" each time the Windows version number goes up, but always at the cost of useful (but messy-looking) interfaces. Quick access to wireless network properties? Too messy! Modify the DDE-commands that the Explorer issues when double-clicking a file? A GUI is too messy for that, users should just use regedit.exe. The "go up"-button in Windows Explorer? Why, "go back" should satisfy all our users' needs!
Of course, destuffying messy interfaces is a noble and important cause. But the degree of mimimalism that current designers seem to strive for is, in my personal opinion, too radical, and I enjoyed reading a piece that points out how sometimes, great interfaces must have lots of buttons.
Yeah I can see where you're coming from here, the Android example does resonate with me. Perhaps I'm not seeing the forefront of this movement because I'm using Linux so much nowadays - I guess I saw the bootstrap example and left it there.
It is odd how we're getting more and more screenspace to play with, phone screens and computer screens are getting larger with higher resolutions regularly, and yet we're pushing to remove functionality in the name of simplicity.
I use it for web side projects. It's not too bad with a theme and some customization, for my simple needs - I'm not a designer, I just want a responsive design out of the box with familiar elements so that a new user can find their way around easily. I can then concentrate on the functionality.
It's boring perhaps but for a certain category of apps you want boring+familiar.
He conveniently overlooks the fact that many of these sites sporting minimalist UI, are doing so due to lack of content. They need a natural way to take up the canvas for a site that is low on content.
Complex problems need complex UI to do them justice, it's true.
But it doesn't change the fact that complex UI is hard for users to learn.
If users only use your software rarely, instead of daily, they will not be capable of learning a complex UI.
If users don't know or think of what they are doing as complex or at least important, they will not be willing to learn a complex UI.
If your software started out with a clean simple UI but added features little by little, and many/most of your users have been with you all along -- that was a lot easier to learn, because they learned it little by little, and don't even realize how complex the thing they learned is.
(This also means if you can hide complexity for beginning users and add it in little by little, this can help it's learnability. But easier said than done -- Microsoft Office's attempts to hide 'less used' menu items is an example of failure that made things worse).
A shopping site with a complicated UI is going to fail miserably -- unless you manage to make a shopping site that convinces users it really does something much much better than the competition, and/or a shopping site your users will use daily.
Photoshop is something users use daily, often as part of their paid job, to do something the users recognize is fairly complex work and find valuable -- AND many of it's users have probably been using it since it was simpler and learning it little by little. Newcomers to photoshop (or gimp), especially those not using it for a paid job, often find it very difficult to learn and give up.
Facebook is something users use daily, for something they find indispensable. It also does an _okay_ job of providing a UI where the users can ignore _most_ of it until they need it (and there are many parts that many users have never found even if they would be useful). It's also got the network effect lockin -- a Facebook competitor with a "better" UI would have trouble competing even if it _did_ manage to come up with a just-as-powerful-but-simpler UI, assuming that were possible.
Craigslist -- I think actually _is_ a clean/simple interface. Better graphic design styling probably wouldn't hurt it (I really don't think it would, in any way, if done well), but it's not complicated.
It's true that powerful functionality is very hard, or sometimes impossible, to do with a clean simple UI. And powerful functionality is great. But it's also true, I think, that no matter how powerful your functionality, users are going to find a complex UI difficult, and they have to have a lot of motivation to learn it.
For a shopping site someone uses only once or occasionally, I am still convinced clean simple UI is always going to 'win' (increase sales, get more sales and users than it's competitors all things being equal, etc).
I am also convinced that you should always _aspire_ to as clean and simple an interface as possible. If you can get powerful functionality _with_ a clean interface, users will always like that better than the same func with a complicated interface. (People _use_ MS Word, but few _like_ it. If someone could figure out how to provide the same func with a cleaner interface, people would go wild for it. It's a big 'if', it's not easy). If you can't figure out how to do that (and it's not easy) -- it is hard to predict whether less powerful cleaner software or more powerful complex software will 'win' with the users in any given case, it just depends.
What really annoys me is the usefulness of ugly color coding. But maybe there are better palette choices than on the avid keyboard: http://www.drted.com/IMG_1225.JPG
Yes, I love having to hit command/control + F on a site like Craigslist. What could possibly be a better design than forcing the user to use their find function because the site is basically unusable otherwise?
Agree with this premise, Users will not necessarily appreciate simplicity over functionality.
It's also worth noting that Facebook, Craigslist, Photoshop, etc are not necessarily optimizing their UI for new users, but rather those users who have been using their software for a while. I imagine that one's design considerations will change based on the prototype of the typical user.
I don't feel that this article has a point. For example if you don't like the ui of Facebook you will still use it because there is no alternative (yet). There are so many factors at play like lack of alternatives you don't take into account. I personally think that Craigslist for example is very well organized. Just because it is not material design it can be good.
Sounds like the change from the old Google maps to the new ones where I can't find anything because it is all hidden away behind unintuitive humburger menus or something similar. Sure I get maybe 10% more screen space, but I have too Google how to send a link almost every time these days.
Yeah, there are big problems with their UX. I don't think the problem is the "10% more space" or clean design, just bad decisions for simple UX concepts. I would expect something to be somewhere logically, but it's not. Or I would press back and go to a page I've never visited. Or I would want to go somewhere by walking, but they show me a car icon..?! oh wait, I need to press the car, then I can change it to walk. Why not show me an icon that say "Go There" if it's an abstraction for many moving types.
> so many factors at play like lack of alternatives you don't take into account
It looks as though there are no real alternatives now but that wasn't the case when Facebook was starting out. It was much smaller than MySpace, and Google's Orkut got bigger faster, for example. Later, it had competition from many other rivals including Google+.
There's only a "lack of alternatives" to Facebook's UI because Facebook already won all those battles.
>There are so many factors at play like lack of alternatives you don't take into account
He already covers that, but it seems you didn't take that section of TFA into account.
He says about how alternatives created to be more minimal, nicer to look etc. fail because they don't offer the same functionality, information density, etc. -- and how that is a necessary tradeoff.
He picked two examples that came from a time where UX/good graphic design didn't matter. They are still popular today because of their incumbent status.
I'd wager that neither application would be as popular if they came out today as our standards have been raised.
Shout out to Basecamp for getting attention for another one of their link-bait-y, contrarian articles though. They are so good at that and it keeps their name out there.
The cliché is a self-taught designer in a minimalist loft in some gentrified inner city, drinking espresso from an expensive vintage machine, obsessively cleaning their wooden slab desk, listening to post-techno or whatever...
...making designs that look like they signify design... with some new or rediscovered modernist sans serif font... matching pastel colors... tasteful transitions... high-resolution photographs that signify urban creative prosumerism...
It's a kind of functional tradition, but as a whiny outsider I'm really getting sick of it. So I like this article because it's just highlighting that there are other values to aspire for, not just these matte white ideals of OCD modernism.
Reminds me of a video from some old RubyConf, I think, after a presentation about how "beauty corresponds to quality", some humanist in the audience asked a thoughtful question like "the type of beauty you've explained seems mostly like a classical or modernist ideal; what about stuff like postmodern notions of beauty, could that be relevant too?"
(Ruby itself is ugly and messy, a weird mix of Perl and Smalltalk... Hacker News is ugly and messy... even the interfaces polished to look simple and clean turn out to be ugly and messy when you look closer, so maybe part of it is just about honesty.)