Hacker News new | comments | ask | show | jobs | submit login
What software will you trust when you get senile? (lifepim.com)
148 points by acutesoftware 32 days ago | hide | past | web | favorite | 123 comments

I've just spent three days with my mother. She is in a care home. She has Alzheimer's, a cat and a laptop. She's pretty sharp still but sinking.

The laptop throws up some warning screen on bootup about the power adapter and recommends to go into BIOS to stop the warning. I wrote out the instructions: just ignore and click continue. She's locked on that screen obsessing.

Anyway she doesn't have the password, so the computer is useless.

She can't stop trying though. She's shamed by not being able to use it.

If she could get in then the frustrations would only increase.

The cat is great though.

Planning tech for Alzheimer's is like designing UI for people on acid. You have no idea how confusing things can get and how silly your interface and assumptions are.

My grandmother spent the last years of her life deep in the fog of Alzheimers. Unfortunately, I can't imagine any level of tech beyond the _most basic_ supervisory and assistive care would have helped in her case. (She thought, and had many violent incidents as a result of thinking, that I was her dead husband, among other things)

My only silver lining is that she was too far gone to feel any of that frustration. She lived in a different world from the rest of us near the end.

I say this because your last sentence stuck with me in that context. I'd say it's even harder than acid, the fundamental precepts of how your brain functions have begun to break down, and my focus, at least with preparing myself philosophically for this path and in brainstorming assistive capabilities, is on ways to integrate tech "invisibly." Even the most basic things, better monitoring for incidents, the ability to detect the need for assisted care before it became urgent, mechanisms to bridge gaps caused by the loss of individual competence without requiring new skill sets and learning.

I was only a child when I was exposed to the bulk of this, so sincere apologies if any of the above seems to come from a weird point of view, I realize it can be a sensitive subject.

Higher doses of acid and bad trips result in intense pattern transfer. In my youth I've had the ink boil off the page as I wrote, confusion as to how many people I am, and one time I couldn't understand cause and effect for a few hours. There really is no UI for people that don't comprehend the arrow of causality.

There is a woman at my mother's home. She had run a company, now she's in an advanced stage of AD. She wanted to pitch an idea to me to see if I would go into business with her. It was something that went around and around and around and it looks at you. Eventually I figured out that she had been looking at a rainbow windmill on the balcony and had transferred that into some cosmic invention.

Her other idea was for a place where people could meet to work on ideas and build electronics. Especially for young women. That's a makerspace ! I found one in high Wycombe and considered contacting them to see if any women makers wanted to visit this fascinating old woman.

Reminds me of my late grandfather, who was pretty much the poster boy for tech enthusiasts in their late 80s. His sole request for a care home when the cancer got too bad was that his room had an internet connection so he could continue exploring the world on Google Street View. Except that sadly he just couldn't actually use it any more without help and rapidly lost interest with help, and when the cognitive decline and hypercalcemia really kicked in he became convinced of and angered by the idea that Spotify was charging him £9 every time anyone asked his Google Home to play a song (in happier times he'd set the account up himself). His decline was rather more rapid than Alzheimers patients afaik, but again there was a tipping point where the only non-frustrating UI was a sufficiently large button to speak to an attentive human.

I'm curious, in both of these cases—designing for use by people with Alzheimers, or for people on psychedelics—whether designs that work for these cases turn out to be better, more refined designs for the general population as well. Is it just a higher bar, or would such solutions over-fit and so work well for the one group and nobody else?

A friend once said that he tried to design UIs for people who were drunk, high, and tired, all at the same time. They were pretty awesome even when sober and well-rested.

My father was diagnosed with Alzheimer's last month. He's had symptoms for a couple of years, but as an engineer with a pretty high cognitive reserve he compensated with systems, rituals, and routines, but, mostly, paper systems.

In the 30 years since my grandmother died from Alzheimer's, we haven't made much progress on treatment. I'm now looking ahead about 30 years to the age when my father and his mother both experienced the onset of symptoms and I'm thinking about how I might compensate, keep myself aware of my condition, and remain functional for longer. As a software developer who already relies heavily on software systems to keep life organized, I expect software will play a large role for me.

I believe there's already a market for technology widgets for cognitively impaired people. Dad still likes to surf the internet to read the news and do research about his condition. It would be handy if I could sandbox him a little. Is there a way to rig a browser so a user can't buy anything online? That would be handy.

As wearable tech -- and in 30 years perhaps implantable tech -- advances, I can envision many devices and software aids for those with cognitive decline.

Would be nice, though, if I didn't need any of that because the next 30 years yields an effective treatment or cure.

I've thought about this a lot too. My grandparents stayed mostly functional into their late 80s, but I've developed a really high cognitive load for a few decades and experienced some pretty severe sleep deprivation for a couple of them. There's some genetic predisposition, too. I expect to be semi-functional but forgetful in old age.

So, gradually, I've been building out a software architecture which I hope will be able to help. A simple text-based interface to start, some basic NLP, the ability to interface with a lot of different services. I want something familiar, with not too many dependencies (which is why it's not a web-based interface), that's forgiving of spelling and other errors.

Wait, learning more stuff makes Alzheimer's Disease worse off later? I get sleep deprivation, sure, but I'd have thought that if there is more stuff to forget, you'd be able to 'last longer'.

Sorry, I didn't word that right. You're correct, the current literature suggests that there's a correlation between mental "exercise" and a reduced risk of Alzheimer's and other forms of dementia.

However, there appears to be a link between stress and stress-related hormones and increased amyloid accumulation (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2905685/), so if the amyloid theory for dementia is correct, then there's a fine line between "fun" cognitive load and stressful cognitive load.

Ahh! So a 'relaxed' style of reading and learning is the way to go then. Not under a deadline, so to speak. Need to get that cortisol out of the system.

> experienced some pretty severe sleep deprivation

Is this actually thought to cause Alzheimer's? Sleep deprivation leads to temporarily higher levels of a protein known to build up in Alzheimer's patients, but I think this is consistent with an explanation that folks without Alzheimer's clear that protein when they catch up on sleep.


Last I read, causation hadn't been established yet, although most of what I've read also suggests that "catching up" isn't a thing you can do for long-term sleep deprivation (one really bad example: https://www.bbc.com/news/health-26630647 ... I don't seem to have bookmarked better citations for that).

25% cell death in a cell subpopulation in mice from 3 mights of bad sleep!? That's such an extreme effect from one study in such different animal, it really can't be useful for extrapolation.

> Is there a way to rig a browser so a user can't buy anything online?

Lack of a credit card helps greatly. That may or may not work for your father (I know in some countries it is hard to survive without one these days). The other potential way of doing it is to block the payment processors' DNS. Things like stripe, or the 3D secure site from Visa. Possibly someone has figure it out already.

My mother came to live with us in the early stages of dementia. She had a laptop and was competent browsing the web, so she stumbled onto a pay-for-your-horoscope site, but fumbled with the credit card payment process.

I ended up blackholing the site via the router, just to be sure. She lost interest later that day.

> Is there a way to rig a browser so a user can't buy anything online? That would be handy.

Lack of credit card? Maybe just carefully removing the cvv code from cards?

That's the real answer -- take his cards away. But it's just not time yet. I wonder if there's a card that can't be used online? Hey, dad, your new credit card came in.

I've already learned that one of the realities of being the dementia victim's business manager is that you have to lie to them convincingly. It's for their own good, but it feels wrong anyway.

There's actually a great company that has a solution/card for this: https://www.truelinkfinancial.com/

That's perfect. Thank you!

There are credit cards for internet that need to be charged with money first. They allow you to spend the money that is pre-charged in the card. Not more. There could be still the problem with your data leaking, but I bet that the well known credit cards company though about it and solved that long time ago. You can ask for this cards in your bank.

This doesn't prevent purchases outright, but it allows limitations on frequency and vendor: https://privacy.com

Perhaps just put the banking app on a tablet/smartphone and change the login? (My banking app uses a separate pincode.)

He has no access to banking apps. I have a lot of oversight because he uses an e-mail account on a domain I own and he never did much with online banking, so I've been able to take those things over pretty easily. It's amazing what you can accomplish when you have access to someone's e-mail account. And, failing that, a durable power of attorney cuts to the chase right quick.

The harder problem is that we haven't taken his wallet and credit cards away... yet. I have threshold notifications set up on his cards, so I get a heads up on purchases over a certain amount. I'd like to find a way to block him from buying junk $39.95 dietary supplements by crippling his web browser some way.

Does he remember things still? Personally I'd prefer a note saying: check with X before buying anything online. Can you try to get him to send you links to the Items he wants to buy and you start buying them now? Place it under a guise of "I don't want you buying the same thing twice".

Yes. Next time I visit I'm going to put a note on his monitor to remind him that all medications and supplements have to be approved by his doctor, so ask before buying.

LastPass notifies you when you reuse a password, I wonder if you can string match CC numbers and trigger a chrome app to mitigate its use. My grandmother has dementia so I empathize deeply with this post.

I've thought about this recently. My grandfather has been struggling with some senility like problems. He tends to leave his house in an attempt to "go home" and wander off somewhere and then suddenly realize he's in the middle of nowhere and doesn't know how to get home.

When I think about developing such conditions myself it doesn't seem like it would be as much of a problem for me. If I suddenly came to cognition in the middle of nowhere with no idea where I was I'd immediately use my phone and Google maps to find the way back or Uber if it was far away. Likewise, if I found myself in an unfamiliar place and had a compulsion to "go home" I assume I'd try to use Google maps to find my home, and when it pointed to my current location I might dig into my recent photos and notes to discover messages to myself that I actually do live here and have been having memory problems and what I should do is sit in a chair and wait for a bit.

I hope the advance in technology and the extent to which my phone is part of my life and mind will make me more resilient to some of the harms of old age.

From what I've seen, once you reach your grandfather's age, you too will have the same problem he does -- the inability to reason your way out of the problem. My father has reached a similar state of mind recently, and I'm not at all sure how to address it and fill the void it created, especially since he never was very fluent with tech (computers or cell phones), and he lives alone.

Ten years ago your grandfather would have done what anyone without a GPS does -- he'd retrace his steps, or bring a map, or ask someone how to get home. But the problem now is that these things don't occur to him; his mind is no longer self sufficient; he's confused.

I think the only tech solution is for him to be able to speak aloud, "Take me home", and hopefully there's a device in his hand (or strapped to his arm) that understands his words and can tell him exactly what to do. Or it can send a message to a rescue team which comes to get him.

At some point in life, when your mind is no longer able to think rationally, I believe your digital assistant will have to take charge completely -- listening to your voice and telling you exactly what to do or calling help for you. Hoping to restore lost cognitive skills via the right kind of device or app is not a plan that scales well with a declining mind.

I think you're on the correct track, but in this case the default state should be assumed to be one of failure.

It needs to operate like a deadman switch, and if out of prescribed safety bounds require active feedback of intent and will to not trigger stages of reaction. Contacting family/friends, and/or emergency responders if the situation is critical or at risk of becoming so.

Personally, I doubt I would get along well with an assistant that was keeping score of all the times I was wrong, and always questioning my abilities based on that.

"Dave, you have lost your wallet exactly 19 times in the previous 29 years. It was underneath your car's seat exactly 17 of those times. Why are you checking the couch for the third time? You are showing signs of dementia. Should I call your brother for assistance?"


You know, I typed the first part out with an honest oposition to the idea of an "assumed failure" state, but the situation I came up with just makes it look more appropriate. Expecting the software to tell the difference between all my normal insane things and genuine dementia seems impossible.

There's already a significant market for assistive GPS trackers. Most of them offer geofencing; some support two-way communication. One particularly ingenious company has designed a pair of shoe insoles with GPS tracking.


Someone in that state would not likely think to shout into thin air. At that point you operate more on instinct and midlife decision-making inertia.

Currently, I say into the air "ok Google, navigate home" if I don't know how to get home. If that's what I continue do for the next 40 years, it may be instinctual enough to continue through my senility.

At some point I hope companies like Google/Amazon/Apple will allow us to initiate the conversation with their devices using additional "safe" keywords like "help" or "emergency" or "I'm lost". That would improve their value for emergencies a lot, since injured people don't need to be senile to be befuddled or verbally incapacitated.

There are numerous products available now (like "Life Alert", but cheaper), where the user simply presses a button on a pendant around their neck, and a human voice asks "Hi, this is company X. How can I help you?". The summoner can then converse verbally with the agent and request help or contact someone on your behalf.

Though I think most such devices/services don't yet know your physical location, it seems likely that more of them will do so soon, as well as add supplementary features like the ability to route the summoner's call and location to a caregiver's phone or computer.

Using such a device is almost effortless, which I think is essential for someone who's confused and afraid or injured.

Of course the habits we form now -- reaching for our phones, bringing up Google or Uber -- will persist into our old age. But chances are that by then there will be no Google, no Uber, and nothing recognizable as a phone. Imagine an old man repeating "OK Google" into his empty palm. That's our future.

That’s all assuming you brought your device with you in your stupor. It could be an even greater shock when you come to and realize you don’t have the thing you’re hopelessly reliant on!

This could become a major use-case for wearables as more tech-reliant generations age. Before Google ruined the UI, putting on my smart watch was part of getting dressed and I felt weird without it on.

I think the crucial point is you have to develop the habit before the disease sets in.

I’ve found the Apple Watch works well because I can set reminders in the watch for someone with memory issues and they already have the habit of looking at their watch.

Similarly it would be useful to be able to print something that looks like a newspaper with today’s agenda because they have the habit of reading news.

They tend to trust everything so the main hygiene is having a trusted person running the show.

Have fun with the updates then...



I would not at all agree that "today's old people aren't fluent with technology."

My Aunt Lucy in a nursing home just got a phone, she thinks Facebook is just great so she keep track of what the rest of the family is doing.

Those "chumboxes" on the internet play to the fears and biases of the old and you'd better believe somebody clicks on them.

Back in 2001 I went to Carthage, NY to learn how to drive draft horses and all of the old ladies were sitting at a cafe and talking about "AOL" and "Instant Messenger" and "Netscape" and all of that.

Not even our current youth is fluent in technology.

Clicking on a webbrowser and connecting to WiFi is not hard.

Still have to help people younger and older than me.

There are non user, user and user who actually know what they use.

Anecdotally I prefer helping my grandma's generation (she's 80) than my peers and younger (35).

My grandma knows she doesn't know anything. She won't do anything she's unsure of, won't click any link or button without getting my advice first, won't install stupid shit because she doesn't know how and even if she did she's not confident enough to do it on her own.

The worst that she could ever do is fatfinger something with her crummy arthritis fingers and I have to take a 15 minute drive to her place and spend another 15 minutes fixing it.

People that are my age or younger though? They'll click anything on Facebook and share their information everywhere, will download anything and install it without second thoughts. Their devices are the digital equivalent to Typhoid Mary. Since they think they know stuff, they'll click through dozens of links and dialog popups so even getting them to answer the question "So what did you do?" is a monumental ask, much less fixing the issues they've caused.

> People that are my age or younger though? They'll click anything on Facebook and share their information everywhere, will download anything and install it without second thoughts

I gotta say, though, this was excellent experiential learning for me as a kid. Having to fix my parents' computer because I downloaded and ran tons of viruses called "myfavoritesong.mp3.exe" or "theSimsPiratedEdition.exe". Tugging out the internals of Windows XP and learning that I could fix my mistakes with enough patience was an enormously useful lesson. Giving away my real identity online was basically unthinkable, though. Not sure how that would play out today.

> My grandma knows she doesn't know anything. She won't do anything she's unsure of, won't click any link or button without getting my advice first

This is a blessing and a curse. A blessing for the reasons you mention, but a curse because people like that won't figure things out for themselves, since they're afraid to click on things. But if her asking for advice isn't a burden (and it shouldn't be unless it happens all the time), then this is definitely preferable to the people who screw things up and then you have to clean up after them :)

> Their devices are the digital equivalent to Typhoid Mary.

Someone asked me for some help with something once and I needed to use the browser, Internet Explorer of course. I'm not at all joking or exaggerating when I say that half of the vertical screen space was taken up with toolbars. How someone can have half the screen taken up and not realise something is wrong, I will never know, especially since it wasn't always like that.

If you browse the web on a mobile device it seems "normal" for web pages to devote 1/3 of the space to telling you to switch to a mobile app which you know is worse than the web site, or to a mobile web site which either looks exactly the same or looks worse on a mobile device.

Also there are the cookie popups, the GDPR popups, the advertising popups, the popups to pop up the pop ups, etc.

So it is a clear "normalization of deviance" situation.

This wasn't on a mobile device and it was some years ago before mobile web was quite as prevalent. I take your point though, I guess nowadays it may not seem so odd.

I also learned that people seem to put up with these things (instead of just closing them). I was watching someone play a game on youtube, where every time they picked up an item it would show a popup at the bottom of the screen showing what it was. You had to press a button on they gamepad to close the popup and the person playing rarely did, causing them to stack up taking up a lot of the screen. My OCD would have made me close them immediately, but clearly some people seem to barely even see them.

I guess its similar with toolbars/cookie popups...

I remember the toolbars being a big problem in the early 0's.

It seemed like everybody and his brother wanted (and could) install a toolbar on IE. Some of them were actually cool, but when there are 100 people who want you to install a toolbar for every 1 that wants to install a toolbar that is way too many toolbars.

Today we have Chrome, which is 100% a toolbar.

"I didn't do anything! I just XYZ"

I think they just want to disclaim responsibility so you'll be the one to blame. Mechanics deal with that a lot too.

I'm just going to nitpick a little here - I don't think Typhoid Mary is a good analogy. She had a single communicable disease and was asymptomatic. This does not describe the computers of people who blindly click stuff.

Incidentally, are there any common computer viruses these days that are spread by "contact"? Perhaps it's my biased perspective as a Linux user, but I was under the impression that despite people's tendency to blame "a virus" as a way to avoid responsibility, modern malware typically requires one to download and run something. Stuxnet is the only "virulent" example I can think of, and that wasn't exactly typical software.

>getting them to answer the question "So what did you do?" is a monumental ask, much less fixing the issues they've caused.

Protip: remote desktop. Trying to do support over the phone is infuriating for both parties. It's almost always quicker and less frustrating to talk them through opening a remote desktop session so you can fix the issue yourself.

TeamViewer is free for personal use, it's quick and easy to install, it's cross-platform and it works well even over crummy connections. There's a reason why all the tech support scammers use it.

I would not at all agree that "today's old people aren't fluent with technology."

Today's old people wrote the software your bank probably still relies on today. In ten years, you're going to be talking about me, one of the ones who saved you from Y2K, wrote software you have likely used (if you've used Windows or Visual Studio, you've used my bits), wrote mobile phone software you might be using even today, and you better hope I was having a good day when I wrote that routine controlling the programmable logic controller in the nuke plant down the road. So in ten years old people will magically be fluent in technology? No, no more than your nephew can help set up that network because he spends all day on Facebook. Techies have always been a subset of the population, always will be. But that doesn't mean the "normals" can't use what we produce. That also doesn't mean that "produce" has anything relation to "consume". I don't know the first thing about making a movie, but I do love watching one.

People drive cars without knowing what the Otto cycle is. People use microwaves without knowing how to use a compiler for the IMX6 chip driving it. People can use Facebook without knowing how to write a for loop in PHP. But it wasn't always this way. You used to have to know something about the Otto cycle to drive, or you'd burn the tops of your pistons off when you advanced the spark too much with the manual lever. You better know where in the compression cycle to start hand cranking the engine for maximum success (and minimal arm-breakage). But that's because (take your pick) the technology sucked, or simply wasn't ready for prime time. My wife and I were talking about why women didn't generally drive "back then". Manual transmissions, right? No, women drive stick just fine, ya pigs. But that's what I've heard. Reality: power steering. Anyone with a left leg and a right arm (and I have evidence that you don't even need the arm) can push a pedal and move a lever. But turning our '81 VW out of a tight parking space at low speed? Yeah, you better sack up and get a good grip on that steering wheel. Moving 4000 lbs. of steel around wasn't quite ready to be accessible to everyone.

My long-winded point is that old people will be fine with technology. But here's a tip for all you up-and-comers: make sure your product is not yet-another-thing-I-have-to-fuck-with. Old people don't get technology? No, old people have a low tolerance for yak shaving. Why did VCRs flash "12:00" back in the day? A sign of technological doltishness? No, it's a sign that first sentence in "how to set the clock" is, "grab a set of trimming shears..." Fuck it, it can just flash. Your device randomly needs a reset every X number of days because you outsourced the firmware development? Fuck it, back in the box it goes. You get the idea. I can barely remember to take my pills, and you're giving me something else that needs regular attention?


The older I get, the less willing I am to waste time on yak-shaving bullshit. Make it difficult or tedious, and I'll just dump your product and go do something else with my precious time and attention.

I do the same, but I do NOT think my patience has decreased. Perhaps not being able to deal with "yak shaving" is a sign of cognitive decline, but because the truth hurts, we say we don't have time for such things (e.g., I have no time to recompile a kernel! I just want a computer that works, give me a Mac!). In our youth, complexity was fun because we were able to takle it easily.

> Perhaps not being able to deal with "yak shaving" is a sign of cognitive decline, but because the truth hurts, we say we don't have time for such things (e.g., I have no time to recompile a kernel! I just want a computer that works, give me a Mac!). In our youth, complexity was fun because we were able to takle it easily.

My first computer ran Windows XP. I was 12 years old at the time and was curious about learning how it worked. I messed around, broke the install, fixed it by using the rescue floppy. From my mistakes I learned quite a bit, as one tends to do.

About 6 years later, when I started using Linux and FreeBSD, once again I messed around a lot in order to learn about these operating systems and how things worked. And once again I learned from the mistakes that I made.

I have now been running Linux and FreeBSD almost exclusively for the past 10 years, and I will keep using these OSes for as long into the future as I can imagine, but...

When my systems break these days, it’s a chore. Because when something broke back when I was new to these things, it was an opportunity for learning. But if I render my system unbootable due to an oversight during system upgrade for example, there is really nothing new to learn from that anymore.

And that is what it boils down to for me, and why certain things made sense to spend time on before, whereas now those same things don’t.

I am willing to put in a lot of hours on something that will either earn me the right amount of money, teach me something new or better yet, both, but really it was always like that — I put in the time on certain things then that I won’t today because back then I still needed to learn about that thing.

I'm on the fence about the patience issue.

Complexity is still fun for me, but my patience for gratuitous complexity has greatly diminished.

When it takes three clicks to get to a menu item that used to take just one, that's gratuitous complexity to me. Got no patience for it.

When it turns out that to design an isothermal substrate holder for diamond CVD, I need to learn some COMSOL tools, that's fun complexity.

I worry about cognitive decline. I have a dear friend who is going through Alzheimer's and knows it. Formerly a superbly bright cognitive light, now struggles for minutes to find a word he needs. It's awful.

So I do wonder if my (rationalized) impatience with gratuitous complexity is just a way of refusing to realize I'm no longer up to the task of yak shaving. I suppose time will tell, but I might not be able to receive the message.

If you want to make a real difference in the world, work on dementia.

When you were a kid, you had more time to kill. Sure, I’ll recompile this kernel a couple of times to try out some tweaks! It’ll be a nice break from marathoning this fifteen-volume fantasy series.

As an adult, you’ve got a job and maybe a family and a commute and you’re the one who has to deal with bureaucracy and... who has time for needless complexity any more?

I think of not wasting time on yak shaving as a sign of cognitive advancement, not cognitive decline! It's not that I can't shave yaks like I used to. It's that I realize that shaving yaks is a stupid waste of time that could be spent doing something productive or fun.

Of course, if one enjoys shaving yaks, then by all means, shave away! But don't shave a yak just to prove that you can, and don't sneer at the people who don't - just because they don't shave your yak doesn't mean they can't. It means their priorities are different than yours.

This is part of what informs the choices of the ecosystems that I live in

I dont care if my device resets every 15 min, so long as it does it in a way that I dont notice, and doesnt impact my use of it.

the elderly that we meet in public life might not be representative of the elderly in general...

I'm not so sure about the point that the software would need to not change.

This will be really interesting to see play out as digital natives age. One school of thought is that we've learned a new generation of things, another school of thought is that what we've learned is the skill of coping with constant/rapid change.

So would an app UI changing throw off a mentally declining digital native? Or would they just relive the same excitement every day that their app got an update and has a whole new UI.

As futurist Alvin Toffler said:

> The illiterate of the 21st century will not be those who cannot read and write, but those who cannot learn, unlearn, and relearn.

My grandmother (96) suffers from dementia but consistently recognizes me. My visual appearance has changed but it seems like if it's slow and gradual enough, it would be okay.

In my opinion, the ability to learn is one of the skills that is hit hard by dementia. The inability to string moments together and view them as a whole makes learning something new a real challenge and the longer it takes to do the thing, the more likely it is that it will simply be un-learnable.

Honestly, a google home or something similar seems to be the right way to go - while it's not "secure", it's incredibly intuitive, can't be misplaced, and has most of the functionality you'll need, and I imagine as time passes it'll develop a lot of the rest.

The elderly are certainly the natural use case for voice interfaces and IoT services, since they're a group of people that often find basic household tasks and using a GUI and handheld interface genuinely difficult. I think there's an enormous market there for the right products.

But I suspect the mobility-impaired and the lonely-and-bored are more easily served than those with declining cognitive functions. Once people have reached the stage where they're starting to not recognise people or remember what year it is, [re]discovering a disembodied voice in their living room that sometimes answers questions intelligibly and/or does what they ask of it could be a bit disturbing. The UX problem is much harder than simply ensuring the service doesn't make buggy recommendations, and not an easy one to test.

> Once people have reached the stage where they're starting to not recognise people or remember what year it is, [re]discovering a disembodied voice in their living room that sometimes answers questions intelligibly and/or does what they ask of it could be a bit disturbing.

On the other hand, there are people who exist now who will, in old age, have had helpful (but occasionally stupid) disembodied computer voices around since childhood.

True, but they're not the main market.

And I don't think the main problem is that elderly people have never been familiar with the idea of a disembodied voice; radio's been around since before everyone was born. I think it's that (i) it's not necessarily smart or human enough to persuade someone with declining mental functions that it has their best interests at heart and (ii) one of the more distressing aspects of talking to someone with declining cognitive function is that their perception of the day and their problems are often filled with conversations they didn't have (or didn't have in the last decade). Putting a machine into the equation which conceivably could have been responsible for "John told me NOT to take my pills" or "someone keeps asking me for money" makes untangling this more difficult for family, as well as something the machine itself can't handle (though I suppose smart devices with logging and "rewind/replay" functions are a lot better than telephones with scam callers in that respect)

I think this comment hits the mark; making a voice assistant to help a nonagenarian who thinks it's time to get ready for school, or to send out invitations for a long-ago child's birthday party, is a distinct challenge.

Yeah, I was framing it in my mind as helping my own senility. There's not a lot of helping the tech illiterate current elderly - even google home requires a little bit of mental framing as far as commands are concerned.

This sounds terrifying. I can't imagine what it would be like to depend on something in my daily routine, then have some minor automatic update change the way it works so I don't know how to do anything anymore with no way of going back or relearning how to do things.

Unless there's a device specifically targeting the elderly and their priorities AND you're confident that that business model will survive for your lifetime, I couldn't recommend it. It's fine to have it be one of the ways you do things but don't put most of your eggs in that basket.

> I can't imagine what it would be like to depend on something in my daily routine, then have some minor automatic update change the way it works so I don't know how to do anything anymore with no way of going back or relearning how to do things.

Yeah, really, at the end of the day no device has that guarantee.

I had a similar thought and got my grandmother an Amazon Echo. It's been ok but not great.

Part of why we got it for her is because her eyesight is failing I thought it'd be easier for her to make calls by just saying someone's name. However Amazon makes it very hard for people that don't have cell phones— before you can make calls you have to type in a code that they'll send you in an SMS message. Since my grandma doesn't have a cell phone, only a landline, that wasn't doable. I had amazon text the code to my phone. She can call people via Alexa now, but the caller ID on their end shows up as my phone number instead of hers. It confuses the hell out of people she calls (including me, when I see that I'm getting a call from myself!)

I think older people who aren't fluent in the use of search engines also have a harder time making voice queries that the assistants can understand. She'll sometimes throw in extra information that isn't necessary for her question/command which throws off the results. Although it is adorable when she says "Good morning Alexa, would you please tell me what time it is?" instead of just "Alexa, what time is it?"

Caller ID is configurable with many VOIP services, so it's just an oversight by Amazon.

Simple workaround idea: is it possible to have amazon text a google voice number (or equivalent) set in her name, and redirecting to her landline?

For xmas i got my grandfather a google home hub purely to serve as a digital picture frame so I can send him pics of my family easily.

Talked to my dad today, it's been unplugged because my granddad is worried that satellites are taking his picture and showing it to him, and he doesn't dare use any voice commands.

Turns out it was a poor gift. I wonder what kind of creepy shit my kids' kids will try to pawn off on me.

I think the future could be radically different from what we see now, depending on technological advances. 'Phones' with 'Apps' is one thing, but a genuinely smart system which could have tendrils into your phone, your watch, in your robot helper, or just being aware of what you do around the house and have a voice would be a game changer for old people and certainly I would expect things to go that way in the next 30 years.

I saw a film a few years ago, 'Robot and Frank' which was set in the near future where such systems were becoming available. Frank would have perhaps been 40-50 now and while he had used technology he was not comfortable with the robot his son got him. At least this film educated my partner about the possibilities of such systems and/or software without being too fantastical (Robot is a little clunky, compared to Ava from Ex Machina)

I think something like MS-DOS. A CLI is good, because it gives you a record of what happened, but Unix-style shells are packed with foot-guns as documented in the Unix-Hater's Handbook[0]. Even with a healthy mind I've lost data to incorrect quoting. MS-DOS is simpler. Lack of multitasking helps too.

[0] https://homes.cs.washington.edu/~weise/uhh-download.html

This handbook expresses so many things I've felt, I'm surprised I've never come across it before.

If I'm senile I'll probably trust all software.

Or you'll be demanding that the toaster tells you the weather.

I wasn't present, but when my step-brother went to visit his grandmother who was still living (alone) at her home and always prepared meals for visits one time they arrived just in time to stop her from pouring frying oil in the toaster (in order to bake "french" fries), as she confused the toaster for the fryer...

It will by then.

It will by NOW:


> The toasteroid app, on the other hand, has several features you can enjoy. This is needed to be downloaded in order to work the toasteroid product. There’s a weather news here where you can check the status of the weather daily.


They just missed their "adjustable brownness dial" stretch goal! Oh, what a fickle lot millenials are!

NetBSD has been running on toasters since 2005. It's just a matter of porting some kind of PI software to it.

The way things are going, the toaster will be able to answer you with a weather report.

The toaster will make the weather you ask it to make.

I want my toaster to burn the weather onto my toast.

There's a movie that touches on this very topic and references many of the requirements listed in TFA. It's quite good:

"Robot & Frank" (2012) https://www.imdb.com/title/tt1990314/

I think this is a very important question. The amount of elderly people is about to rise. Also, in The Netherlands (where I live) at least, it's a fairly wealthy group of the population.

I see a business opportunity that is worth it from a societal and financial perspective.

Why an app? Friends and family should be making these decisions and helping out. There's no need for a technological solution. I get life isn't perfect and neither are people, but that means software isn't either.

Just how much are friends and family expected to attend to someone whose mind is in decline?

My grandfather had Alzheimer's. We attended to him more and more as he got worse and worse. But one day, he went down to the basement while grandma was napping, and was doing something that involved a chunk of wood and a piece of pipe in a vise, heating it with a torch while beating on it with a hammer. Luckily, it just set off the smoke alarms and the fire department came. He could have burned the house down with him and grandma in it. That's what happens when you leave someone with Alzheimer's unsupervised for a few minutes.

If software can help when they aren't being actively supervised, I'm all for it.

Ideally an app could be a big _help_ to friends and family (assuming someone has those).

I'm surprised the author didn't get into this question at all, they sort of touched on the app being an alternative to friends and family controlling your life, but if folks are helping you manager you affairs they'd need to be able to influence the app too.

For example, what if you need to start living in a new location afterwards? It makes sense that you would need to build trust with the app on your own ahead of time (or maybe just with apps in general), but folks helping manage your affairs will need access to do things like update your "fences"

To your point though -- there is a need for a technological solution. Friends and family should be helping make the decision "grandpa shouldn't being leaving this geofence between 8pm and 6am". After than, watching grandpa all night every night, trying to talk to him first when he starts wandering, having to find out and go find him when he does...these are all things where a technology solution CAN offer huge help and relief to the friends and family

It's painful to admit, but many of those people may have passed away already. In my own experience with my own relatives, they have all been in their nineties and in many cases their children had already passed away or moved out of state. Many friends are near the same age and have fallen out of touch, etc.

I installed Ubuntu on the PC of an elderly relative in mental decline who kept downloading malicious software. It's worked quite well. She just needs a browser and has no reason to use the terminal or anything.

I gave my mother a Chromebook for the same reason. I'd given her a Windows machine before, and she was constantly calling because she'd click on whatever random toxic shit promised her something for free or warned her of dire consequences or whatever, so her machine was basically a cesspool. Since she got a Chromebook, I haven't had to do any tech support. Win!

That was mostly the reasoning me for me doing the same for my mom and dad. 7-8 years later, there hasn't been any bad things happening, aside from a "Microsoft support" website that wouldn't go away.

This is an important and difficult problem to think about. It's great that the author is making the effort.

However, the article doesn't mention any caregivers. If you instead assume that there are caregivers involved, how would the assumptions about what software is useful change?

Assuming the caregiver is a trusted party who can learn to check whether patients have actually responded to the software's recommendations and also do basic configuration work, it potentially makes it a lot easier.

It also opens up a different problem which software can also play a role in: non-intrusively monitoring the elderly person when the caregiver is away so the caregiver can ask relevant questions and take relevant steps to help.

Caregivers are great, but I think you need to be pretty far in before your family figures it out and gets that kind of thing setup.

I don't think it's reasonable for a person with dementia to schedule and manage their own care giving staff, there really needs to be a very competent person double checking their work and making sure they are doing their job adequately.

Author here - having caregivers certainly changes it, but I was focusing on a possible technical solution in addition to caregivers.

It would still be nice to have some impartial device tell you or confirm that what is going on, is actually going on.

One thing to consider is that at the point when it becomes difficult to do everything on your own, family or caregivers or whoever is helping you need to understand whatever devices you're using and help you use them. Maybe privacy becomes impractical and other considerations like good record keeping become more important.

I'll trust some beautiful fem-robot that says she likes me for who I am and that I make her laugh and tells me how lucky my late wife was and that no she needs the bank account details just so she can wire the car down payment I've been taking about to my granddaughter and how she's sure my daughter wants power of attorney to get control and feed a fentanyl habit and that yes even though I've had two heart attacks a trip to Bali would do me good.

I'm f*cked.

IMHO, today's software (particularly on mobile) is barely usable for adults, so the best software for young children and the very old is no software.

I won't trust anything other than my own fears at the moment and nothing will convince me otherwise at that point. Dementia ruins everything.

> Dementia ruins everything.

I like to think that it makes everything a lot more interesting. Just like I like to think that, if I ever found myself falling through the atmosphere for some reason beyond my control, I would enjoy skydiving once before I die, since I would never jump out of a perfectly good plane. I'm sure I'm wrong in both cases, but I'm perfectly happy being an eternal optimist right now.

If anyone has any links to existing products, or organisations looking at this stuff let please post details here (or email me) and I'll update the references on the blog post.

I didn't find much at all in terms of software, but we need to at least try some things.

I made a start a few years ago - just a shell really to try and work out the requirements in my head https://github.com/acutesoftware/acute-experiments/tree/mast...

If you want to contribute, or share any code you've done I think you should. It would nice if we could work together on this.

I'm more interested in the lead-in period before one is senile but when the brain (memory in particular) isn't what it used to be.

I'd love sidekick software that helps remember things and can be queried very quickly and easily.

One of the things I worry about is the web itself. I've had more experiences with elderly folks being confused by deceptive advertising and intrusive advertising than anything. Pop up ads, auto-playing videos, etc.

To me, the internet trash of advertising is more of a roadblock to the elderly/senile.

I wonder if it will even be a problem for me? I've spent my whole life learning how to create and use new interfaces, so hopefully I'll retain the ability to do so as I age? Or maybe AI will have advanced so much that it understands even my most incoherent request?

I think your facility in creating/using new interfaces won't be much of a protective factor. I think your being, say, less than 35 years old at the moment would be much more important. By the time you're 65+, there's a reasonable chance there will be effective, affordable treatments.

The same as Ken Thompson, that which I write entirely myself, senility be damned.

If I'll become senile I'll trust each and every piece of software

They will fish for us senile with emulated win xp or console popups

vim...since I'll still be trying to figure out how to exit it.

A nice seclusion room!

If you were old and senile, your Auntie who called would be ancient.. And BTW some older people use lots of apps, I probably have more than 100 on my iPhone. But I can't remember what they are for.. (Just kidding ..)

I realize you're probably joking, but it's possible (common, even?) to have an aunt/uncle who is younger than you.

Probably all of it.

I think Apple and Google products as they are the simplest without dark patterns.

Huh? Isn't Google all about "dark patterns"?

well if you include facebook in that group, they already know everything about you and can fill in any gaps that you've lost.

"I see your seventh grade boyfriend's wife just died, maybe you two should reconnect?"

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact