Losing our product to button syndrome | Hacker News Reader

Losing our product to button syndrome

olickel.com

128 pointshrishi108 comments

It's not just button syndrome. Lots of companies continuously fail their users in terms of just being to able to perform basic functionality while marketing themselves as world-changing, good-doing champions for people.

Being a developer, I'm one of the main points of contact in my family whenever a relative can't figure out how to do something software-related, and boy has it been a sobering look into the future.

I recently helped a relative file for unemployment verification through ID.me, which is a popular identity verification platform. My relative, who is not well off financially, had an old phone that didn't play nicely at all with the ID.me verification flow. I spent an hour trying to get my relative signed up and I never could get it to work. The site was barely mobile-friendly, and the photo upload process kept failing, which was a required step for verification.

It was so Orwellian to see this kind of UX on a device that wasn't new (and of course, how is someone on unemployment expected to purchase a new phone?) I truly wonder how many people have starved because they didn't have access to devices that allowed them to collect their unemployment through this platform. It really kept me up that night.

That isn't realistically fixable unless companies are really willing to invest a lot more in testing. Most companies don't even have any user testing for products beyond what devs do on their own machines.

I realize that, it just put the entire software industry into perspective for me. We are building products for users that can afford to interface with them, and even then, it's not a guarantee that you're gonna get a great experience.

Here's to hoping that neither you or I ever become that irrelevant, because that doesn't look like a pretty existence to me.

For a government service, testing of older/lower performance devices should be part of the initial contract.

It often is, but will be tested to a "yes, it is technically possible" level.

How about devs who say "oh, its IE, its only 1% so we don't care"? What is the point of testing something that you know is not supported anyways?

They invest a bit in the initial development then call it a day. Another example is the chatbots fad: after the technology implementation there's almost no energy for actually training those bots, so you as user are served with a fancy and overpriced menu system.

Chatbots serve a real purpose: they hide information and frustrate users so that they will not request refunds and such.

They are also a stepping stone to real support personnel, and a subset of users fall off the stepping stone making support cheaper.

Which is why we need user-bots, that represent their clients interest with the same emotionless, ceaseless interest as a cooperation would represent its interests.

We're guilty of this, despite caring about the product and the users. It's hard to focus on maintenance and deep iteration when there's more critical (and cool) functionality to chase. Not to mention that iterative UX improvements have the same problem as negative results in academia - it's hard to get people excited about it.

I had the exact same experience. A barely smart phone being used to verify ID and we had to do the facial recognition more than 10 ten times, each time having to go through a whole step by step process to get there. And it still didn't work. I think he had to go an entirely different route in order to even do it. If he didn't have someone like me who could think systematically, there's no way he by himself or any non-technically minded person would be able to do it.

What does facial recognition have to do with it?

You had to take a picture of your face in order to match it with your ID or something like that. We took the picture ten different times in varying light conditions inside and out, and I don't know if the phone was too old or what but it always came back to an error.

To be fair, public libraries generally have computers suitable for this purpose.

For that you have to be living in proximity to one and be able to afford the time and money to get there.

Access to government services should be a right accessible to everyone, not just the select few that can jump through intentional hoops!

Are public libraries not readily accessible to everyone? I thought that was one of the few government services still running fairly well.

Every town I’ve lived in has had at least one well run library if not a network

I live centrally, in my town. The nearest library is an hour walk. The bus would be at least 30 minutes. I'd have to know beforehand that I need to reserve a computer before using it, and that spots fill up quick. I'd need to get a library card to reserve the computer, which I need to do in person. I'd need the bus schedules, and presumably my device might not be capable of accessing that or I might not be knowledgeable enough. I'd need to have cash on hand to pay the bus fare, so I'd need to find an ATM or convenience store that offers cashback.

This labyrinthian effort is basically what being a poor person is like, on a daily basis.

I haven't gotten a library card in a long time, but I think you need proof of a permanent address in the same jurisdiction that funds the library, so if you're homeless, you can't use it (though homeless often use libraries to sleep and wash in, even if they can't use the actual library services).

No library I've ever been in has required anything for someone to walk in and read or use the computers.

My local library requires a card to use the computers, with the exception of the ones to search the database and one public machine.

That being said, the librarian's there are quite helpful to those who need assistance with anything like that - going ao far as to ait with people and help them navigate it.

> It was so Orwellian to see this kind of UX on a device that wasn't new (and of course, how is someone on unemployment expected to purchase a new phone?) I truly wonder how many people have starved because they didn't have access to devices that allowed them to collect their unemployment through this platform. It really kept me up that night.

It may be disheartening to hear, but this is by design. A (Western) government cannot get away with entirely not providing or dismantling basic elements of a social safety net (unemployment insurance, healthcare insurance, disability assistance) for publicity reasons... but what perfectly works is to make the system as complex and hoop-jumping-dependent as possible to reduce the number of claims:

- requiring modern devices (or not making sure that older devices work too, like you witnessed) is a major hurdle many people who are too poor and under-served by libraries or other public Internet access

- requiring in-person presence with short opening hours during weekdays discriminate against people who have to take care of sick relatives/children, have to work two or more jobs or have certain mental health issues that make following up with appointments very difficult (e.g. some of the strains of autism)

- requiring specific forms of ID or other paper documents (e.g. birth certificates) can be almost impossible (or, very expensive) to solve for people who have lost their belongings/are homeless

- requiring proof of residence is an automatic exclusion of the homeless

- complex forms with bureaucratic language discriminate against illiterate people, non-English speakers and frankly, most people who don't know or can't afford a lawyer to help them out

- automatically rejecting the first claim and only allowing after an appeal / a lawsuit is commonplace for disability claims, it is very effective in "weeding out" poor and already troubled people

The ones that do not require a lot of bureaucracy are not governments... the void that helps those left behind by governmental bureaucracy is more often than not religiously affiliated: churches, Salvation Army, other charities - but unlike government (which is theoretically bound by constitutions and anti-discrimination laws), they are free to choose whom they help and how much.

And now: good luck if you're a publicly outed LGBT member, a person of color or otherwise marginalized person right in a religious-conservative stronghold. The government won't help you as you can't jump the hoops that were designed to be that way, and the "private sector" won't help you as you are not mainstream conformant. This threat is what makes dysfunctional government bureaucracy so insidious.

Well, that's super cynical.

The less cynical take is so many people abuse the system that we have put in anti abuse protection. However the end result is if you can figure out the system you are abusing it as anyone who can figure it out can get a real job and care for themselves

> The less cynical take is so many people abuse the system that we have put in anti abuse protection. However the end result is if you can figure out the system you are abusing it as anyone who can figure it out can get a real job and care for themselves

This is even more cynical, and just flat out wrong. Just because you have succesfully navigated the system doesn't matter you can hold down a permanent full time job. Also, "so many people abuse the system" is just nonsense. FullFact [0] estimates that £2bn was lost to benefit fraud in the UK, some of which is reclaimed after the fact. This is about 1% of the benefit budget, or 0.01% of the UK's budget, before any of the reclaimed amounts.

[0] https://fullfact.org/online/comparing-benefit-and-tax-fraud/

And depressingly accurate.

I've seen and tried to help people who are navigating these systems, and this describes my observations each time.

I live in Texas fwiw.

UK's Conservative Party once specifically went about ordering a system so it would fail certain percentage of people, then threw Atos (the contractor) under the bus when it came out, IIRC.

Literally UK government policy under Theresa May: guidance notes to the home office were leaked that instruct the bureaucracy to create a "climate of hostility".

I'd say it is super realistic

can totaly relate from experience; last example was my wife trying to signup for our municipalies self-serving portal (...service?) and beeing utterly frustrated about roadblocks/error-messages and unhelpful service personal. it often looks like nobody is even trying anymore.

thou, shifting the "navigation of service meshes" away from buerocatics, law literacy and ppl-skills to proficiency with electronic devices has done little for the kafkaesqueness.

From your name I'd guess you are a fellow German... I can relate with what you say, too :'D

> it often looks like nobody is even trying anymore.

The key thing behind that is diffusion of responsibility: everyone knows that shit is broken beyond belief, but since there are so many different layers of responsibility, politicians (with whom the responsibility should rest!) often are successful in deflecting liability in a lot of different ways. The end result is that nothing changes.

In order to verify my identity to access gov.uk services, the app complained that my Android 10 phone was too old (a curt "update to a newer version" was the only error message visible). It's already mental that you have to own a smartphone with google spyware to access government services (what about people who don't own such devices or don't know how to operate them?), but now even the second most recent version of android is not good enough.

I eventually borrowed a friend's phone. It proceeded to stonewall me and insist I was not a real person, again with no visible error message or explanation apart from a strange oscillating blue 3D head... Kafkaesque doesn't begin to describe it.

It's indeed one of the weirdest modern tech phenomena: you are blocked/flagged/shadowbanned/what have you by an inscrutable AI and given no reason or explanation, just a blank "you have been banned".

Speaking of, whatever happened to that federated(?) ID thing where you are given the choice of several companies like Experian to validate your identity? I signed up years ago to do one thing and never used it since.

Always seemed a bit odd to me, because I would have thought that even if the government didn't trust its own departments to validate government-issued documents securely (fair enough, actually, they'd probably leave it on a train), all they had to do is provide their own separate, secure, service. They say the government won't know who you're validating with, but why is that an issue anyway. At some point you have to trust the government if you're verifying your identity with their own documents in order to access government services.

Though I guess it's probably just because when they eventually do get hacked they can just blame the provider and ditch them publicly rather than have another IT disaster in the papers with their name on it.

I believe you are referring to the gov.uk Verify service [1]. It's still in existence, but on it's way out due to poor uptake.

It was pushed agressively at one point by GDS, but despite what it claims about being easy to use, it provided a poor user experience, with very high failure rates for certain demographics.

Universal login, which was the vision for this product for UK government services, is a seemingly appealing proposition but if a high percentage of people can't use it, it's pretty much doomed.

The various UK government departments adopted it sparingly or developed alternatives, which further undermined it.

1 - https://www.gov.uk/government/publications/introducing-govuk...

I've used the UK verify service, and it was pretty horrible then. Only country I've seen do it somewhat well is Singapore, with Singpass.

My husband was not able to download his earnings report from the social security website because the verification system did not work for him. My guess is it's because he doesn't have a recent credit history (we've been debt free for five years now) but it worked fine for me so who knows?

The other option was to go in person to the local social security office, but at the time it was completely closed due to COVID.

It’s gross, and all because of the inability of Congress to allow a rational identity framework and the investments made to make it difficult for poor and minority people to vote.

State Departments of Labor have access to extensive consumer data and are unable to utilize it effectively. ID.me shouldn’t exist.

I write guides for immigrants. It's eye-opening to see the sort of problems they face in a system that doesn't account for them.

People are often missing a lot of inherited context, like what the steps of the process are (for buying, insuring and registering a car), what the terms mean (all the different numbers and IDs).

Then it gets messier. Immigrants have to get everything at once: papers, permits, certificates, IDs, accounts. All of them depend on each other, and most of them take weeks to get. People need an income now. A big part of my work was to disentangle a few catch-22 situations around residence permits, bank accounts, apartments and health insurance.

I've talked to a few employees at the immigration office, and since all of them are native to this country, they seemed surprised that a whole Facebook community was dedicated to navigating their office and its requirements.

The same problems apply to tech, but it's worsened by the fact that the builders are well-paid people in developed Western countries, but the users can be anyone, anywhere.

I moved from Ireland to the UK - both ~are~ were members of the EU, and there was no visa requirement to move. The only thing I had to do when I got here was get a bank account and a national insurance number (think SSN) for tax calculations. I tried 6 high street banks (RBS, TSB, Santander, Barclays, Halifax and Nationwide) who would not give me a bank account without a utility bill for proof of address, and would not accept my tenancy agreement, however I couldn't get a utility bill without a direct debit. In the end, someone told me HSBC would give me a "passport" account, and I used that to bootstrap the rest of the process. It was an absolutely infuriating process full of "computer says no", and absolutely nobody willing to help.

I had a similar experience, and I had a passport to prove my UK citizenship. It's a giant knot where everything relies on something else and you need to keep at it until one thing let's you in. It felt impossible and I feel like I had the best case scenario; I can't imagine how people without a support network can do it.

I think it's worth talking about a couple downsides of a search-based interface:

Speed: Search-based puts a floor on how fast an interaction can be. "Button Syndrome" interfaces are a slow "hunt and peck" for new users, but experienced users can use them extremely fast, building up muscle memory so they don't even need to consciously think about the action they're taking. Imagine piloting an F35 with a search-based interface.

Discoverability: Buttons make it explicit what functionality is offered by the product... somewhat. Users may not know exactly what a button does, but they can make an educated guess based on labeling and context, and it gives them a jumping off point to experiment with it or find it in the documentation. With search-base interfaces, there's no natural way for a user to discover functionality they aren't aware of. Worse, a user may remember a function exists but forget the terminology, flailing in the search box guessing different terms.

This is not to say search-based interfaces are bad. There are mitigations to the downsides (the article mentions a few, like search suggestions), plenty of upsides to go along with it, and let's not pretend that button-based interfaces are all sunshine and rainbows. I only mean to say that these are things that should be considered.

I think the broader takeaway of the article is: Always be thinking holistically. It's important to consider how your users interact with your product as a whole, not just the individual features. Also important to consider how different users of varying experience with the product and the domain will feel with the UI—Often features for "power users" come at the expense of new users or vice versa.

I'd add data collection/privacy concerns that list too. "search suggestions" seem great until you see that often every letter you type is being sent over the wire for analytics and to mine for personal info to sell to 3rd parties. The site even mentions how useful it is for them to see what people are searching for. I don't know anyone who hasn't ever typed or pasted data somewhere they didn't mean to and it's painfully easy for sensitive data to get leaked this way.

That concern won't apply everywhere, but it's worth keeping in mind. Even when my interactions with a simple menu are recorded (for analytics/profit) and reported the amount of data they get is at least limited.

My favorite compromise is a search that takes a user right to the button, and or can initiate whatever it is.

Each search is an opportunity to build experience needed to use search less.

The buttons are there for those who want to run fast and or efficiently.

Doing that well is a lot of work, but it also delivers high value.

Going search only can be super lean, which has to be compelling. Everything costs something though, and the cost here is no user becomes adept. There is a permanently fairly high Ccost of interaction.

> My favorite compromise is a search that takes a user right to the button, and or can initiate whatever it is.

This is how MacOS X's help menu search box has worked for over a decade, and it's brilliant. Type in a search and it shows you every menu item that matches, and rolling down the list shows you live where each menu item is.

I wish Spotlight search was similar - showing paths more readily, and making it easier to open the folder that contains the item that you want to look at.

I don't know how you're supposed to discover this, but highlighting a file result in Spotlight and holding the Command key will show you where it is, and Command+Enter will open the parent directory in Finder with the file highlighted

That's the problem. Not very discoverable, but pretty efficient for those in the know.

Goddamn. I've been an MacOS user, and supported MacOS users, for ages and I had no idea this existed. Thank you!

I first encountered that method in NX CAD.

Siemens is an old company who tends to get investments right. Apple uses Siemens design software, so who knows? Perhaps Apple had influence?

Great implementation!

At the time, I was in a heavy pre-sales role and would make the help system a part of the demo. I am pretty sure everyone saw high value there.

And talk about complex, with depth, subtlety! That program demands more than mere buttons can deliver!

I first encountered that method in NX CAD.

Siemens is an old company who tends to get investments right. Apple uses Siemens design software, so who knows? Perhaps Apple had influence?

Great implementation!

At the time, I was in a heavy pre-sales role and would make the help system a part of the demo. I am pretty sure everyone saw high value there.

And talk about complex, with depth, subtlety! That program demands more than mere buttons can deliver!

And my own ability to use the tool improved dramatically. I have 20 plus years doing sales, service, and support. OTC, SDRC, UGS (Both those two now Siemens, Soludworks, Solid Edge...)

Hands down, these methods crush anything else I have seen.

Some of our inspiration definitely comes from CAD, NX CAD and Fusion. IMO the work they've put into making a mountain of functionality useful and intuitive to a new user has paid off.

And that's the functionality in front and planned to be favored by users. Many older features are available to those who know how to ask for them.

Also, in terms of the geometry, how the system builds models, the feature version is stored with the features, and that version will build models in newer software, until migrated forward by a user.

It's very complex.

Thanks for this! We're definitely working it into our system. It's always good to have multiple paths when possible, so a menu that connects to search and backwards might be quite useful to introduce users to features, but tie in muscle memory once they are used to it.

I use VS Code, which offers a keybinding interface. But when I use the command search, the cost is really low, because most of my searches require zero, one, or two characters to disambiguate.

Why not both? Reaper has actions you can search for, but you can also put them in custom toolbars. I've been getting into Reaktor lately. It's like a virtual modular thing, but ranges from low-level (programming DSP) to high (playing instruments or plugging effects together). The main way to find things to add when patching stuff together is via search. You can create macros, basically little containers with inputs and outputs, and pull them into new projects and wire stuff into them.

I would not be surprised if newer fighter interfaces are already customized for the pilot in this way.

I loathe "smart" interfaces that want to cram everything into a dynamically-adjusting control panel. Because what the programmer thinks I want to do is so frequently not what I actually want to do.

Give me panels that I can open, close, and arrange as I see fit. Let me craft the interface of your complex, featureful program into something that matches what I need most often. Pick a simple set to expose to new users that covers the basics, sure, great idea. But let the skilled user tell the program exactly what sets of knobs they want to have handy.

Every program needs windows/linux-like start menu and explorer/finder-like sidebar. This ui idiom is so successful and familiar that it's just stupid to not make it into a program. We need to stop designers who think only colors, fonts and transparent 40px frames.

Author uses aircraft cockpits as an example of "improvement".

I've been getting back into DCS recently, and one thing that's struck me with the A-10C (II) is just how incredibly resilient the entire design is.

Sure there's a lot of work starting the aircraft and setting up systems etc, and there's an immense learning curve, however you have a staggering amount of redundancy in case of failures, in all levels of the aircraft design.

As a consequence, pilots have been able to limp home ridiculously damaged planes, surviving to fight another day and go back to their children.

Then there's the F-16. I've only just begun to scratch the surface of this one, the simulation is equally as amazing as the A-10C. This aircraft is likewise very manual, but if something goes wrong the pilot has immediate control; in the F-35 (from what I've read) he would be completely reliant on the glass/ computers working as intended / desired, with a level of complexity that makes Tesla's entire FSD program appear as basic calculus.

That's perhaps a big tangent to computing, however I strongly feel that simplifying everything might take away more than you add. The best designs would combine quick, express usage functionality with expert panels for advanced, expert users/usage. This should be combined with a transparency of design that allows their users to understand the underpinnings.

That's a good point at the end, and where we'd like to be. Unfortunately dev time is always a finite resource, especially for an early-stage product. You need to pick one over the other - at least a little.

> Unfortunately dev time is always a finite resource

true, but usually not the limiting factor.

more to the point; money is a finite resource if the product is intended to generate a profit.

The OP doesn't actually describe what he thinks is wrong.

> It isn't just us, here's Office 2003.

> [image]

> Here it is after the rebuild.

> [image that is similar, but on a narrower screen]

> To understand why - and to find a solution - the history of software is helpful.

Step back. To understand why what? To find a solution to what?

And the screenshot isn’t even from MS Office 2003 but from … Kingsoft Office.

I was curious so I searched Google. That specific image appears on the second row for me. I suspect some carelessness on the author's part and trust in Google's ability to return appropriate results. Though, to be fair to Google, the text underneath is "Download Microsoft Office 2003 Look...". The "..." expands to "-Like Kingsoft Office Suite - AskVG" when clicking on the image. So the text suggests that it is an appropriate image, and probably by some SEO and popularity metric it gets a place near the top in the results.

I found that confusing too, like I walked into the middle of an existing conversation.

Ironic that the OP commits the same sin as the bad examples that he's describing - namely, neglecting to take into account the user's potential lack of context for the issue at hand.

I thought so, too. Especially, since the Office 2007 image is so small, that I could not get anything out of it.

If there were a science of user interaction, its second law could be called the Wide Angle Fallacy. When a disgusted user goes back to the designer saying, “Your system doesn’t perform the special function I need,” the designer’s ego is deeply affected. To regain the good graces of his customer—and to re-establish his self-esteem—the designer is likely to answer, “I can fix it in no time. I will just add another command for you.”

Later, the same man will be seen at conventions, meetings and workshops, extolling the virtues of his system, the “power” of which can be measured by the great number of commands it can execute. I believe this is usually a fallacy and users should recognize it as such.

— Jacques Vallee, The Network Revolution: Confessions of a Computer Scientist (1982) chapter six, Obfuscatology (https://books.google.com/books?id=6f8VqnZaPQwC)

No software I can think of has been ruined for me because of getting too advanced.

To particular pieces of software has lost a lot of their utility for me thanks to dumbification:

- Google I have now given up. It was equally dumb as DDG when I last used it and the only reason I sometimes fell back to it was to see if it randomly provided a useful result.

- Firefox is still the best for me but is a shadow of its former self. I'm eagerly waiting for a fork and on Mac I have already switched to Orion which has built in vertical tabs, can fix ctrl-tab and support both FF and Chrome extensions. (My main criteria is: 1. works great 2. not Chrome- or Chromium-based)

Bloat and feature creep are real (and I'd accuse Firefox of both) but in general yes, I'd rather have more functionality and a complex interface than have features stripped away or hidden behind some search box where I have to know exactly what I want and hope the app can guess at what I'm asking for.

Agree to a large degree, but this:

> Bloat and feature creep are real (and I'd accuse Firefox of both) but in general yes[...]

Except Pocket, what particular new bloat has Firefox added over the last decade?

I am not too fond of Mozilla myself, but if anything I feel like Firefox has been stripped down way to far.

I totally agree. I have had a LOT of complaints about Firefox in the past five years and all of them fall into one of two buckets.

A) Useful functionality being removed. Firefox on Android has been crippled to the point of ruin.

B) Pointless UX changes for the worse. This can be bearable, but are a regular source of annoyance and wasted time.

Pocket is what comes to mind immediately absolutely, but there's a long and growing list of things I have to disable every time I install Firefox including:

- Their entire ad filled start pages / Snippets / Activity Stream / Suggested sites

- Normandy

- safebrowsing

- Push notifications

- Firefox Feature Recommendations

- Firefox Hello

- Firefox Accounts / Sync / Firefox Monitor / CloudServices

- Shield opt-out studies / Test Pilot / Experiments

- pre-installed 3rd party search engines

- PDF reader

- screenshots@mozilla.org.xpi, pictureinpicture@mozilla.org.xpi

- Password generator

Don't get me wrong, I'm a fan. It's still the most customizable browser available and with a lot of work (and a few add-ons) the most secure and private by far, but lean browsers don't require all this to prevent unsolicited network connections : https://support.mozilla.org/en-US/kb/how-stop-firefox-making...

That was a longer list than I expected, thanks for putting in the effort and also for your nuanced position.

I'm one of those who enjoy the PDF reader FWIW, same with push notifications. The PDF reader in particular because I almost haven't used Adobe since it came.

Test pilot is fine with me as long as it is voluntary and opt in - but the robot extension - that I literally felt in my body when I saw it.

Again, thanks.

I think most of this should have been created using an extension API and should been voluntary to install. Back in the days this would absolutely have been possible I think.

No problem! I know a lot of people love the PDF reader, it really is convenient, and I'm a big fan of avoiding adobe's reader whenever possible but after so many years of issues the file format is tainted to me. Too many exploitable features and having popular browsers include that functionality makes those readers an attractive target.

I don't think Mozilla has done a comparatively terrible job with it (they all end up with vulnerabilities, and it would take a lot to steal that crown from acrobat) but I do feel better downloading PDFs and using something a little less common to view them. I've bounced around between using Open/LibreOffice, Foxit (before they went bad), and even converting PDFs into saner formats prior to viewing, but these days I'm liking Sumatra for windows

There's a distinction to be made between "capable but not overwhelming", "capable and overwhelming", "incapable but not overwhelming", and "incapable but still overwhelming". The first two can both describe advanced systems without dumbification, but the user experience is qualitatively different and can lead to wildly different outcomes when put into use.

You can have an "advanced" (whatever it may mean in context) system which fits into the first two categories, which is very useful to remember. Something I like is command prompts in the style of emacs accessed via M-x and similar shortcuts (or in VS Code, which many more people are familiar with). These permit discovery of new commands and activation of commands without overwhelming the UI. They can also "teach" the user, by providing information like what the keyboard shortcut actually is for activating it. Contrast this with something I've seen in many desktop projects (especially ones targeting a smaller number of power users, versus a more public system distributed to a broader user base): menu hell. All those same commands are still there (maybe), but buried in menus with submenus with submenus. Even though a command may logically appear in multiple places, it probably only appears in one. They may not even appear in a logical place, but just a conventional one, like search commands showing up under "Edit".

I am not so sure that replacing physical fixed function buttons with context sensitive buttons in a fighter is really a good idea.

If you look at the Navy, there have been several mishaps that have been blamed on poor electronic controls.

In addition, and likely more importantly, a fighter pilot trains for hundreds of hours in their cockpit. They develop muscle memory. Having a button at the same place, with the same feel, that does the same thing, is likely vital when you are engaged with an enemy fighter and don't want any extraneous distractions. Instead, it seems you have the cockpit version of Apple's Touch Bar.

Heck, I can't even consistently hit the correct touchpad-buttons on my Honda's radio. I can't imagine trying to do the same in stressful combat situations (both physically - high Gs and speeds - and mentally).

I know this isn't the point of the article but the supposed F16 cockpit in this article is an F15 cockpit. The F16 has a sidestick, mfds and the canopy has no support arch.

Also I believe the F-35 still has a landing gear lever, I would never want my gear lever to be contextually defined. I want that visual cue as I scan my cockpit prior to landing.

The screenshot is taken from the video talk he linked further up, and is mentioned as an F15 in that video, so it's a typo.

Good catch - fixed!

But I like buttons. The more, the better. Complexity is more a function of depth than breadth.

I don't trust your integrated search feature to understand what I want, and in any event, I rarely feel like playing a game of Zork to get my work done.

Please use buttons, clearly labeled in English (if not the user's native language.)

Looking at your screenshots, you pretty much had it right the first time.

I like buttons too. Quite a fair bit. But when you get to too many buttons, they become icons.

Add that to the fact that all user tests we've done have fared better with search and more verbose interfaces, and I wonder if we are both of us far from the target user profile.

On my iPhone 6 with safari I’m just seeing an error message:

> Application error: a client-side exception has occurred.

But from the other comments I get the impression that “button syndrome” has something to do with UX/accessibility. Tad ironic.

A lot of times too many buttons/UI is just lack of design. I can't say if that's the case here as well, but I've seen it many times.

Usually well intentioned founders who know what they want to achieve, but fail to understand the subtleties of designing UI's and interaction. Or developers who sometimes lack the viewpoint of a "regular" user.

In most cases hiring a good (interaction/ux) designer can solve this pretty quickly.

Maybe a bit pretentious but it does kind of state it nicely: "To complicate is simple to simplify is complicated. Everybody is able to complicate. Only a few can simplify." - Bruno Munari

He missed one of Office’s key innovations - commands-on-selection, which is now available on the web app as well.

It reinforces the story even better than the given examples and has been there since Office 2007. The web version just added search-for-command in the right-click menu as well, which is similarly powerful.

This seems to be called the “mini toolbar”. Very hard to Google, easier to find through use. It is in Outlook for web now, too. The version in word is more powerful, especially when working with tables.

What is “commands-on-selection“ ?

I googled it - but that only led back here!

If you hover your cursor over certain commands in the ribbon, you'll see a preview of the action in the editing area. This was one of the major feature of Office 2007.

That wasn’t the feature I was thinking of, but it was indeed one of Office 2007’s major innovations!

That release got a lot of grief for folks used to menus, but it really was a major leap in UX. The discoverability advantages were huge. Another favorite: rich tooltips with images and text descriptions for commands.

Sibling comment is a better description of what I was referring to, specifically floating the most common formatting commands next to the cursor on selection. It saves a mouse round trip to the ribbon.

Eg when you select text, a table, etc. in Word it provides a toolbar of popular commands (font, layout, etc)

I think I want to agree with this article, but I found it a little confusing.

In particular, I think it's strange to compare the F-16 with the F-35 when the former is regarded to be one of the best fighter jets ever made and the F-35 is infamous for being problematic. My understanding is that they are also different kinds of planes for different purposes, but at any rate, I struggle to focus on UX when there's that contextual elephant in the room.

> In particular, I think it's strange to compare the F-16 with the F-35 when the former is regarded to be one of the best fighter jets ever made and the F-35 is infamous for being problematic. My understanding is that they are also different kinds of planes for different purposes, but at any rate, I struggle to focus on UX when there's that contextual elephant in the room.

Also bringing up fighter jets in general might not be the best comparison. A fighter pilot is about the farthest thing from a casual user of a fighter jet. My understanding is they're intensely dedicated to knowing how to operate it effectively, to the point of muscle-memorizing menu navigation button-presses to quickly perform different actions (e.g. memorizing something like the Konami Code to fire a particular type of missile).

So they're almost certainly not like the users of whatever web app your building, so maybe it's not the best idea to let a photo of their controls inspire your UX.

[1] https://en.wikipedia.org/wiki/Konami_Code

Or even if they chose a Tesla without knobs and buttons, people complain about that endlessly. Sure everything is in a screen which is flexible and easy to program, but nobody can figure out how to adjust the heat.

I agree. Not having flown a fighter jet, myself, it's not clear to me which I would prefer. I can easily imagine problems with the "simpler" F-35 cockpit requiring the pilot to navigate a modal UI while navigating the aircraft.

Even older aircraft like the F-16 and F-15 have many modal UIs. They also have a lot of UI features that were added on over the decades which make the system much tighter to sit in than the original pilots would have experienced. The big difference between them and an F-35 (or even F-22) is that the newer ones have the opportunity to clean up or unify a lot of those UI systems so it feels less ad hoc (at present, give it 30 years).

The modals are usually for things you're not going to use often (comparatively), or are very low "depth" menu for modes that have major impacts on what you're doing (so you're essentially selecting activity type)

A bit more problematic is that it uses a photo of a fighter jet that is definitely not F-16, it appears to be early F-15A?

Interesting article. It looks like the solution you came up with works well for your product which is the most important thing.

I see a lot of products these days using more of a hybrid approach. Some of the most basic buttons to get new users used to the product and discovering what it can do. Then adding in a CMD/CTRL + K search menu for when users reach a better understanding of the product and just need to get shit done.

There is a fine line between too basic and too advanced (depending on the audience for your product) and a challenging thing for most UXers is balancing new user onboarding against the power users of your product.

A friend of mine has been in software development for over 10 years - possibly closer to 20 but I'm not sure on that point. We chat a few times a month about work stuff, and routinely he will express amazement at being disappointed about something not working 'correctly'.

"Well, the documentation says this takes key X, but it doesn't work if I send key X, I have to send Y"

"This service seems to break when this JSON comes back down a second time".

"I press this button and it shows it being clicked, but it just locks up for 5-10 seconds before finishing the click. That's a really bad user experience. I'm surprised companyX doesn't seem to understand that's confusing to people."

And on and on. After... years of these observations, I pushed back a bit with "I'm surprised you continue to be surprised by any of this. You've worked in software for years now. You should probably be surprised that anything works and continues to work."

A system continuing to work between various client/browser versions and system upgrades over a period of years, serving multitudes of users, with as little downtime and no data corruption is really... surprising in these modern times. It shouldn't be, but there's so many moving parts, so many players, that long functioning services do surprise me.

A recent bank merger meant that two regional banks are merging. A recent mortgage refinance meant that this newly merged bank also bought my mortgage, and I'm supposed to make payments to them. Their system for setting up a mortgage payment is broken. Like.... errors in dev tools console, with relatively obvious root cause as in a field like CustomerId is referenced in another part of some code as CustomerID, and when I try to click X to move forward there's breaking JS code.

This was a problem for at least 5 weeks that I know of, but from the little I could see via twitter and forums, weeks before that. There is 0 way of me contacting anyone who could even understand my issue in the first place, and that's a whole other topic. That they likely didn't have monitoring in place (I couldn't see any obvious tools such as sentry installed) is another story, but... I didn't chose this bank. I don't have a choice in using them (same with regional utility companies, etc).

The beauty of the Office 2003 type of UI is that is 100% user configurable.

I haven't really felt in control of software since.

Thanks for all the feedback! I've added a section at the end addressing as many as I can.

Lost me at... this is an F16 ( cockpit was certainly NOT an F16 ).

focus!

An excellent article

also a lot of fluff. Lenghty tirade about flight controls and linking a 1-hour flight lesson didn't help brevity/clarity.

pour a drink, sit back and have a read

Captain, it's Wednesday.

Brevity is not the only goal in writing. If the author had not provided an analogue in another field, it would not have been engaging.

I'm not sure I saw the basic yin/yang tension explicitly described that I would like to see.

Back in the 80s, the earliest Mac interface guidelines said something to the effect of "eschew modes". There were lots of reasons for that. Some argue that as more people are familiar with computers, it's reasonable for the standards of good design to change.

But. Gigantic numbers of buttons are kind of a consquence of avoiding modes.

I don't think I need to reiterate all the bad things about modes, because quite a few were mentioned in the article, including life and death situations with military aircraft.

Quote: "segment our interface into a much larger number of smaller pages, each of which serves a specific function" - those are modes!

There's no right answer, and the article covers a lot of the ground, but I think the end of it is unbalanced, because it's fundamentally about a duality with no resolution, where intelligent people have argued for the opposite of the final advice.

There's nothing really wrong with the article, except it avoids the keyword that connects to significant history that shows both sides have merit. "Mode" is not used once.

The older I get, the more depressing it is when I see someone rediscovering something without recognizing it.

Reply on news.ycombinator.com