back
484 comments
Danny o'brien, he runs one of the oldest surviving blogs, oblomovka, coined a term "hinternet" sometime in 2007, that was when the internet was still being run by the technological elite, for themselves, but normal people have also joined. The idea of hinternet was that there was essentially two internets. One is the sophisticated technology and a value add, and the other one is the internet of the viagra pills and popup banners. We, the technology elite, would rarely venture into the hinternet, like going into a bad neighborhood, where's normal people had no such mechanism for discernment, so their experience of the internet was distinctly different and inferior.

Now most of the internet is hinternet, and we're all forced more and more to rely on it. Banking systems, mortgage platforms, car payments, utilities payments are generally designed mobile first, desktop later, they employ various dark techniques for "verifying real user", which break on open platforms, forcing you to access them from iPads and other such locked down devices, or not at all. If hinternet used to be the dark shady streets where hucksters were peddling you knockoff watches, then now hinternet is the dystopian landscape of vertical information integration, ran, behind the scenes, by para-governmental institutions. You can't log in into irs without using id.me, a digital wallet and identity management platform, that sells you things.

There are attempts to cultivate little gardens of sophistication, but they are of mixed success. On a personal level there's a strong disincentive to participate in the hinternet beyond the mandatory, carefully navigating poorly designed and conceived systems just long enough to achieve an objective. One has to login into irs, but one doesn't really need to read that popup and upsell blocked, mobile centric news article.

From this perspective "mobile-first web design" is a symptom removed from its greater context.

This "hinternet" is a cool concept, but there's something missing from its account.

On one side we have the cultured elites of academia, the military and government - as rightful founders.

On the other, the unwashed masses, immigrants of the Eternal September. Eventually this hoi polloi of hucksters, chancers and grifters became naturalised as the businesses and bankers in the new world.

The dotcom era is a colonisation story and the elites are the aboriginal natives driven off their own land. It sure fits a "woke" narrative.

But what's missing from this fairy-tale is the actual real people.

The truth is, dotcom, Web2.0 and the empire building between 1997 and about 2010 was still a marginal affair, where existing money and power moved into the internet, along with a handful of rugged "entrepreneurs" (as we like to call ourselves around here).

The 99% remained spectators caught between the Scylla and Charibdes, and now they are corralled into ranches, all lovingly watched over...

The potential for a "people's internet" still remains, but we have not solved many (indeed any) of the classical problems of freeloaders, tragedy of the commons.... and at this point I think "Web 3.0 and blockchain web" is dead (?)

A good start to moving things forward to an internet that is once again public, high-quality and large might be looking more closely at the history/narrative of the internet and who the real stakeholders are.

I'm struggling to understand from your explanation what is this "hinternet".

The hard-to-use internet of the 90s, centered around IRC and Usenet? The seedy parts of the 90s internet with illegal content hosted on free hosts? Because you say today's internet is like that, and I don't see any comparison at all however I look at it.

How is the 90s Internet (forums, IRC, Viagra spam and goatse) anything like the modern sterilized version full of dark patterns in the hands of a dozen megacorps?

It’s basically what computer illiterates (which is growing, not declining because of mobile use) endure on windows desktop computers. At some time they’ve clicked the wrong link, managed to install a toolbar in their browser - and slowly it’s been infested with random dark patterns. I see it fairly frequently when relatives call me about computer issues.
You’re lucky if it works with an iPad and that it doesn’t make you sign your life away to Verizon or T-Mobile the ur-Carrier (sic).

I alway get voted down when I use words like “phonish” or “phonishness” but I feel that smartphones made life worse not better and made people serve computers than the other way around. Here’s to the next platform.

This is an interesting thing. Honestly can't remember the last time I saw an ad or easy phishing thing. Like obv in your junk mail but thats literally the extent of my exposure to it, everything else is curated and "good" or probably aligns with your concept of the priveleged netizens/areas even though Im poor as hell. Knowledge wise I suppose I rich so there's that but there's also seems like a tradgedy of the commons type situation that depends on all the tech-illiterates to be the meat shields for advertising and paying for things like YouTube
This is the difference between building a website as an experiment for an untested audience and running a modern business. The simple reason most websites are optimized for mobile is because most users are on mobile. And the reason there are ads and email acquisition forms is because they are worth something to businesses. Text-only, ad-free, non-responsive text content isn't worth much.
> There are attempts to cultivate little gardens of sophistication, but they are of mixed success.

Mastodon comes to mind, whose openness allows me to browse it from emacs with mastodon.el.

Removing full or even useful content from both RSS and notification emails comes to mind as well.

Link to the piece if anyone finds this in the future: https://www.oblomovka.com/wp/2003/04/15/the-hinternet/

(It was both strangely broken in my WP database, and also I think my website was down when OP write this. Sorry!)

> You can't log in into irs without using id.me, a digital wallet and identity management platform, that sells you things.

Wow, I just assumed it was some kind of auth flow the government runs themselves and never did any research

Over the years it's become clear to me that frontend is probably the hardest part of the stack. People think they can just shit out some bootstrap react app and it's perfect, but being able to write complex UIs that can work on any browser, any device, with all assistive technologies, and all languages, is extremely hard. You need someone with a deep knowledge of html, css, and the supportive web apis. A good frontend engineer is incredibly rare, even at big tech. What's even more rare is a UX designer that also thinks about these things, who are worth a million bucks.
The problem sadly isn't even mobile-first vs. desktop, but rather designers who still haven't figured out that the web is dynamic content that should be allowed to flow according to the user's display device size and shape. It's not and never has been a static medium like paper. It's not limited to a specific size and shape like paper, and should not be treated as if it were. Web "designers" should not be trying to force the content into any specific size or pixel resolution, as there's just too many different resolutions of screens and width vs. height layouts of those screens to ever be able to cover them all appropriately without adapting to the idea that the content must be able to flow accordingly. It also severely harms accessibility for folks with vision issues who might scale up their fonts to compensate if doing so causes the content to break in horrible ways that make it unreadable.
Hamburgers and Hieroglyphs! Ahhhhhhhhhhh!

I get why they are used on mobile. If you only have room for content, it makes sense to tuck actions away into a hamburger menu except for a small number that you assign to tiny little hieroglyphs. Fine. However, if you have space, this is a terrible way to use it. At best it adds steps, at worst it invites experimentation and disaster to figure out what the heiroglyphs do (which wouldn't be so bad if undo worked but we've apparently decided undo is fine to break too). Like the Apple HIGs used to say, on Desktop you should want to get the most common actions out of menus and onto labeled buttons so that users can answer "what can I do?" without playing hide and seek. Undo should be baked in from the very start (it's hard to retrofit) to reduce the consequences of experimentation.

Unfortunately mobile design has taken over so completely that even on apps which will be used almost entirely on desktop, even on apps with an internal advocate for Desktop design, UI designers go for the hamburgers and heiroglyphs and broken undo because it's standard these days. Sigh.

Oh, and modals are back with a vengeance, but I need to stop here or my blood pressure is going to get unhealthy.

This study doesn't make any practical sense. These pages weren't designed to convey the maximal amount of information in the least amount of space, they were designed to sell a product. It's impossible to claim if these designs have a negative impact due to content dispersion or not unless you are measuring them against the purpose they were designed for.

They explicitly studied ecommerce/product pages here. The relevant metrics are which page had a higher perceived product value? Which page had a higher conversion ratio? Which page resulted in a higher NPS? Which page created a more positive brand affinity?

You don't sell portable speakers using specs, you sell it with aspirational images of it being used on a beach. Of course expanding an accordion of product details then asking "On a scale from 1-7, How well do you feel you understood the offering communicated on the page?" results in a higher survey score. If you said the more dense page converted better, then I would be surprised.

It's like designing a study on the negative impact of hard F1 race car seats, adding a bunch of foam, testing which is more comfortable, then proclaiming one is better than the other because it was rated more comfortable, when the only metric they were designed for is lap time.

Many websites will completely alter their UI based on resizing your window. Sometimes I want to set a browser to take up half of a monitor so that I can put something else next to it. This often fails on modern websites because the UI becomes unusable after the window is resized
Yes! I hate websites that put useful content in a thin column in the center.

What's worse, formerly normal websites start degrading. Typically, when a new manager is hired and decides to "reimagine" the product with the "mobile-first" vision. The recent Patreon is a good example.

And of course, in some cases normal websites go away entirely, and are replaced with crapps: Venmo, Amazon Alexa, Chamberlain, etc.

The thing that I truly find awkward is no nice touch equivalent to hover.

It's such a useful piece of the UX to have thrown away in the move to mobile first.

Other than that I probably fall in the "it's not that hard" camp. Of all the problems you have to solve, getting it to look reasonable on a few different screen sizes is pretty far down the list in terms of time and complexity.

As a webdesigner myself I kind of disagree. Simple content per screen just works better for the example they gave. It’s easier to visually parse. Their condensed version has a lot of multi column layouts which I really dislike.

High content density works for desktop applications, but not for what’s basically a brochure website.

I’m less concerned with brochure and ecommerce sites. But mobile first design for productivity sites and tool interfaces drive me crazy. Who normally accesses these tools from their phones?
What you really want is two apps, completely different UX's and its not worth the effort more times than not. No one is going to Herman Miller's website for example and not buying a chair because they don't like the desktop web app experience.

They might pass though if they can't get the site to work well on their phone.

I hate the “hero” UX concept and how everyone uses it in a useless landing page. Scrolling and clicking through endless marketing BS to figure out what you do and why it’s relevant to me. It’s the business equivalent of restaurant sites that don’t put their hours, address, and phone right in your face.
It's not just web design. Desktop software is being ruined by mobile-only app design.
I enjoyed making simple, fast websites using HTML, CSS and a smattering of JS for progressive enhancement, usually with Django on the backend.

Industry got obsessed with JS frameworks made for giganto-orgs as a “solution” for everything.

Eyes became fixated on the claustrophobic mobile-first screens.

Went back to doing primarily graphic design. Less money, much happier.

Mobile-first is not mobile only design!

This article has completely misunderstand the term, mobile first it's the technical way of organizing the CSS to render the mobile viewport first and then the rest of the 'breakpoints'.

Mobile-first design has the advantage, over the desktop-first design, that it directly render first the mobile design and the cellphone experience is much faster and with less rearragning flickering.

Mobile-first doesn't have the corresponding disadvantages at the desktop browser since they tend to have a land line internet and much faster cpu and memory.

Omg yes!

It has finally been told! I was crying and talking about making the return of information rich websites, but people were just following trends blindly :(

I’ve been a web developer since before “mobile first”, and since before mobile as a serious web target. Granted when it did become a serious target, we had different terms as well: adaptive and responsive design come to mind. “Mobile first” as a concept was not then—and never has been, for anyone who takes those other concepts seriously—“mobile at the expense of all else”.

What I mean is that quite a lot of these examples and others people frequently cite when complaining about “mobile first” are not inherent to “mobile first” per se. To my mind, they’re an incomplete application of the principle. And the principle became prominent when the inverse problem was more universal: designs (or simply their implementation) targeted desktop first, and added mobile affordances as an afterthought.

“Mobile first” shouldn’t mean that other web experiences aren’t just as important a consideration. Philosophically, it comes from the perspective that a broadly usable and accessible web experience accounts for the most stringent constraints and works out from there.

I distinctly recall solving problems like those discussed in the article well over a decade ago. It was a lot of work. It requires a lot of care and attention to detail. That doesn’t excuse skipping any of it! And it really should be more achievable as the standards have evolved. But it does require dedication to addressing a large matrix of users’ needs and usage conditions.

The irritating thing is that it is neither mobile first nor desktop first. Most websites I come across are designed and built for desktop but using a mobile aesthetic.
> to consume all content.

I've decided, whenever anyone weighs in on information/interaction design, and says "consume...content", what they are saying is, "Hey, we've been stabbing everyone in the face all wrong, here's a better way to stab everyone in the face."

Vanguard has been guilty of this as well. Its recent site redesign has been all over the place, and sometimes leaning heavily toward folks using its phone app (not that its phone app is very user friendly either; for example, checking what orders one has placed by account is impossible). It used to be that the website has a table view that shows almost everything I need to know about my holdings. Now, it takes a few clicks to find that info and even then, the layout is so sparse (mobile optimized) and hard to read.

Maybe I'm getting really old and just like to complain about this because I'm not very much used to this phone-oriented UIs.

The visual design problem mirrors a problem with the organization producing the website. Before mobile, a website was the entire online presence of a company; every subdivision within the company had a presence above the fold. The culture and conceptual structure of web development developed around this concept. As mobile mandates pushed things below the fold or off the site entirely, web teams found themselves unprepared and have adapted awkwardly, as this article shows. Mobile app teams never had this cultural baggage, which is one of the many reasons native apps are perceived as superior.
It’s an indicator that most of their userbase is mobile. Look at B2B SaaS apps and they tend to have the opposite problem
Shit like this makes me want to learn more about using accessibility features of my browsers. I'm far from going blind, but I don't want to scroll through the angry fruit salad of alegria art, huge stock images that take a huge amount of screen real estate and the important bits being tucked away probably somewhere in the bottom.

"Minimalist design" and it still takes multiple megabytes of bandwidth to show a pretty nothing.

I'm using Reader Mode or an equivalent whenever I can. If a site cannot be dealt with using RM, I mostly go elsewhere.

Now get off my lawn, kids.

This is definitely not a web-only issue. I have seen the same issue with convergent KDE apps (Qt UI based).

Not an expert in this space but intuitively it feels that what is missing in a widely adopted heuristic of how to "best", or at least, "least-bad" map the information and interactivity components of a given UI to different screen geometries.

It feels like a hard problem to crack: in some cases there might not be a good solution at all. So the heuristic should be able to say "no, what you are trying to do does not match any known Homo Sapiens sensory configurations".

Think e.g. how you would map a large and busy DAW screen [1] with hundreds of dials on a small mobile. You'd probably have to rethink the entire workflow, segment into small chunks, smoothly switch screens while providing summaries and connecting elements etc.

[1] https://soundcamp.org/music-and-sound-production-blog/best-f...

The apps go where the money is, most normal people are mobile-first or mobile-only.
Skimming through the paper format article I struggled to find any grounding. I was curious to see hard numbers such as time to parse information, instead I only see anecdotes like "one participant said".

Not that I agree or disagree with the proposition, just not enough information to form opinion.

I've been long advocating to get rid of the mobile-first approach and instead embrace a width-agnostic paradigm.

It is possible to ensure that a div or image doesn't stretch too much on desktop by setting the right max-width. It is possible to ensure that a tabular div doesn't overflow on mobile by setting the right width in a responsive @media query.

The problem is that mobile-first has become mobile-only, assuming that 80% will come from those devices and the others will figure out a way to cope with the eyesore. Real responsive design and craft of width is still possible (now like in 2010), but it takes time and testing - resources that companies are more than willing to save, since everybody in tech seems bent on pushing half-baked MVPs nowadays and call them products.

This is a really long article to say "Content dispersion is bad".

Yeah, obviously. More so if you use a generic image like Waze. It's not a usability problem, it's a quality problem. Apple does just fine with full width desktop heroes because they include relevant images.

One mobile-first cancer that's becoming increasingly annoying is sites highjacking Ctrl/Shift/Cmd+clicking (used to open links in new tabs).

I paused writing this comment/rant to do a quick search about it and found that someone had already done what I intended to do - write an extension/addon to deal with it:

https://github.com/danielnixon/link-fixer

Thank you Daniel Nixon!

Not directly related (or is it?): I just hate sites where all I see is some fancy wallpaper image and a slogan (if at all).

Then I have to scroll down on my 27" monitor to be able to actually read anything on that site. That is stupid - at least in my opinion.

My favourite dead-end task is when I'm working on some established non-mobile first UI that has often been haphazardly put together with only 13" laptops in mind, and asked to essentially "make it responsive". It's a tiresome conversation to have when the task bleeds into another sprint, usually with a surprised manager who I'm sure could have finished in 1 hour between meetings. Don't forget the tests, QA, managing other PR reviews, participating in pointless meetings, and inane last-minute requirements changes
> Usability issues of dispersed content include:... > Increased cognitive load > Difficulty building a conceptual model for the page

This completely agrees with my own experience. But if these things are true when viewing a "dispersed content" site on a desktop browser, shouldn't they be just as true on mobile? It seems to me that this is a big problem with mobile-first design in general: it's just not the right format for actually conveying conceptual information to users.

I thought mobile-first web design meant that you start by designing for mobile and then adding features for the bigger screens upward, not to prioritize mobile screens.
I've been seeing this more in content, and also apps that are "browsers" really - some view control that have a mobile & PC/MacOS equivalent - like they are intentionally formatting everything for portrait type display on mobile

My devices are a literal ton of iOS ones, and then 4k and a 2560 × 1440 display which I use most frequently lately.

Tons of whitespace, oversized text everywhere and incorrect use of Z-index are some of the hallmarks of this mobile-first design.

Yeah, as someone who can't use touch screens, this is a serious pain point in a lot of my web browsing. Mobile first design is an accessibility problem.
I disagree with almost all of these conclusions. I appreciate web pages that are sparse and concise. High information density is _not_ generally a good thing.

I like clean websites with lots of space. I don’t mind long scrolling websites, though I don’t particularly like the trend of stupid animations that happen as you scroll down.

I appreciate that “mobile first” has forced web designers to simplify web sites to be concise. Less is more!

I could say the same about web on desktop. Desktop native is still the best for desktop. Browsers insert friction and are rent seekers.
Sorry, desktop, you're gonna have to suck hind tit. Most users are on mobile, so that's what gets targeted.

Even open source is like this now: GNOME, Wayland, and systemd are designed around the most common use cases their developers anticipate and if you fall into one of the more obscure categories... well, tough.

PREACH

opposite way in: 2013 article about high density on the japanese web being (in part) a legacy of their pre smartphone era

Think a lot of the issue is the difficulty grasping what size the actual screen is. I get webpages on my 1080p monitor where everything is just far too big but I guess as far as the code is concerned it has limited knowledge of whether the physical size of the screen is 24" or 5"