This is one of the few HN articles that have profoundly moved me. Such a beautiful and simple use of technology to make a clear and big improvement in someone's life.
As a side note on his mother remembering that the tablet exists, it sounds like she has amnesia quite like Henry Molaison, a famous case study in neuropathology. He had very specific brain damage that seemingly stopped him forming new memories in the same way as OP's mother, but studies showed that he could remember some things, just not consciously. So for example he would have warm feelings towards people who'd been caring for him despite not remembering them, and would also pick up card games more and more quickly as he played them repeatedly despite saying he didn't remember the game. OP's mother remembering the tablet sounds very similar, particularly when paired with the feeling of being remembered and loved by her children.
> but studies showed that he could remember some things, just not consciously.
This reminds me of muscle memory. I can play pieces on the piano even though I don't actively remember the sheet music of them. My hands just "know" what to do. Funnily enough the moment I start actively thinking about certain passages that ability worsens by a lot.
At the start it's all about carrying around notes full of picking the relevant condition depending on the current permutation/state of the cube then following the step by step algorithms on which sequence of steps to perform for that condition.
Then you'll naturally realise that certain conditions happen a lot more than others and you'll start to remember the sequence of letters for each series of steps to perform.
Over time you'll forget the letters and your fingers will just know the sequence to perform when you perceive that condition, kind of like typing a password without thinking about it.
Eventually you'll be able to fit each condition and algorithm into your muscle memory and completely forget the series of letters that you used to memorise.
Now I can barely explain how to solve a rubik's cube in-person. I just do it.
Yes same for me on guitar. If I try to play something too slowly or if I really start thinking about what I'm doing it all falls apart.
I think that's when you really know a piece, when you can play it incredibly slowly. Paradoxically it's easy to play quickly and just let your fingers play out their muscle memory, playing something really slowly is the challenge.
I ran into this when teaching my son to tie his shoes. He now ties his shoes “upside down” from me, because I tied it from my perspective. It’s surprisingly hard to tie shoes in slow motion, it took some practice by paying attention to myself tying shoes quickly.
Now I’m wondering if you can tell a kid is from an “even” or “odd” generation by which way they tie shoes…
I remember a lecturer in undergrad psychology talking about this in the context of walking, and my walking felt really messy for a week, like when you start to become conscious of your breathing.
In psychology memory is divided up into various groupings depending on what people are interested in, e.g. explicit (remembering that Paris is the capital of France) and implicit (remembering how to ride a bike). You can further subdivide explicit into semantic (Paris is the capital of France) and episodic (events that you have experienced), and implicit into procedural (how to ride a bike) and emotional conditioning (memories of feelings). Those categories aren't related to neurophysiology though, which is where I think it gets really interesting because I doubt matches those rather Platonic categories.
Further than just muscle memory, every cell in our bodies actually has "memories". That's why heart transplant patients can experience personality changes from the donor:
Excuse my ignorance in asking, but is this trustworthy? I'm a layperson regarding biology and I was always assumed that organs outside of the brain don't contribute to memory. At the end of the article is the statement "Data not available / No data was used for the research described in the article." Is it possible to see the data?
We know there are lots of biological mechanisms that retain state at the cellular level to put it in CS-ish terms. A fraction of these mechanisms could plausibly be transmitted outside the cell (e.g., miRNA).
These mechanisms may or may not encode memories as we typically understand them, i.e., the ability to remember an event or fact, but could very plausibly shift personality, preferences, etc.
Not to mention that most neurotransmitters are produced / collected from the gut. Many seem to be produced / used as signalling molecules by gut microbiota.
>> can experience personality changes from the donor
> organs outside of the brain don't contribute to memory
Interesting question. To start, personality typically refers to the totality of a person's behaviors, not the memories they may be able to bring forth. Behavior, esp automatic, is informed by cognitive states informed by the body.
Affect is the general sense of feeling that you experience throughout each day. It is not emotion but a much simpler feeling with two features. The first is how pleasant or unpleasant you feel, which scientists call valence. . . . The second feature of affect is how calm or agitated you feel, which is called arousal. [0]
Simple pleasant and unpleasant feelings come from an ongoing process inside you called interoception. Interoception is your brain’s representation of all sensations from your internal organs and tissues, the hormones in your blood, and your immune system.
...[M]oment-to-moment interoception infuses us with affect, which we then use as evidence about the world. People like to say that seeing is believing, but affective realism demonstrates that believing is seeing.
0. Barrett, Lisa Feldman. How Emotions Are Made: The Secret Life of the Brain (p. 72). HarperCollins. Kindle Edition.
Yes it is strange to practice a song one day and then come back to it again the next day. It's like meeting a new person who plays better than I did yesterday, and practice involves finding out more about this new person.
My choir director does this with new rehearsal pieces on purpose. We go through them once at the beginning and then let them "percolate" while we practice some other songs. Then we go back to them in "stabilization" before the end of the same rehearsal and they suddenly feel familiar, so we can pay better attention to things like dynamics. It's wild.
> but studies showed that he could remember some things, just not consciously
I expect it is very hard to overestimate how incorrect our mental model memory and learning is. If literally everything was forgotten, then you could set up a reverse groundhog day or groundhog hour for someone, just optimize for them having a wonderful day every day. (Would still be horrible for the loved ones to be effectively disconnected from their still-living relative.) Probably there have been movies made about this.
I have no experience with this but I am sure it is nothing, nothing, nothing like that. The article says you wouldn't wish it on your worst enemy.
> Because she cannot remember things, she goes through each day in a state of low-grade anxiety about where her grown children are and whether they are all right. She feels she hasn’t heard from any of us in a long time.
To me this is not a description of someone frozen in time. To me this is a description of some horrific combination of some amount of learning or "remembering" happening, some sense of passage of time, and no episodic memories to draw on to explain any of it.
> If literally everything was forgotten, then you could set up a reverse groundhog day or groundhog hour for someone, just optimize for them having a wonderful day every day. (Would still be horrible for the loved ones to be effectively disconnected from their still-living relative.) Probably there have been movies made about this.
There is a Drew Barrymore movie Fifty first dates. And yes, it is horrible for the relatives.
Perhaps if we approach technology more from the perspective of elders, and those in need, we are going to produce much better technology application for everyone else.
There was a study that suggested that the motor cortex can remember even if short term memory conversion was destroyed.
If nothing else, myelinization counts as a form of memory. Strengthened by reuse.
I would love to know if those warm feelings are stronger with individuals who remind you of someone you used to know. “This nurse reminds me of Aunt Sarah, who was nice to me when my dog died.” And so forth.
I have this weird issue where about a third of people I meet for the first time swear they know me from somewhere, and it's somewhere specific that I know I've never been. My dad and brother have the same issue, and we strongly resemble each other, so I think I just have a congenitally familiar face.
I have no idea if feelings would automatically transfer to me from people with amnesia, but they certainly do for people without it, even though I don't remind them of anyone they know well enough to name.
That study is an interesting suggestion that there might be a physiological basis for the explicit / implicit distinction in terms of memory. Makes sense in many ways that some kind of memory might be embedded in the motor cortex. I wonder if the same is true for emotional memories and midbrain structures, as hinted at in your last paragraph.
I always find those non-obvious connections fascinating, like the disorders where e.g. someone can't say the word "fork" when they're looking at one despite being to describe what you use it for etc, but can immediately name it when they touch it.
What a beautiful use of technology to uphold someone's personhood, and let them know they are loved, despite (and with regard to) a profound injury.
This reminds me of a desire I've had for a long time: a simple, wall-mountable eInk device that could be configured with a URL (+wifi creds) and render a markdown file, refreshing once every hour or so. It would be so useful for so many applications – I'm a parish priest and so I could use it to let people know what events are on, if a service is cancelled, the current prayer list, ... the applications would be endless. I'd definitely pay a couple of hundred dollars per device for a solid version of such a thing, if it could be mounted and then recharged every month or two.
For anyone else that followed the "buy a device" link on the docs page, and found yourself on the (ended) Kickstarter page, editing the URL to https://usetrmnl.com/ works :)
(This is fantastic. Thank you for sharing about it!)
> Most IoT products support SSH-ing directly into peripheral devices. We've heard too many horror stories about how this can go wrong, and decided to invert the paradigm.
> Your TRMNL device pings our server, never the other way around.
> Each request made to our /api/display endpoint includes only the minimum details needed to support customers -- an API key, device mac address, firmware version, battery voltage, and wifi signal strength.
Super hackable but it pings their hosted server and nothing else?! Is there a way to run your own server?
we're adding more docs on running your own server soon, which will include 1-click deploy starter projects that Just Work.
if you think about it, we are incentivized to do this. no subscription fees means the more you ping our server, the lower our margin. but for now we're wrapping up fulfilling all pre-orders, scaling, etc typical new product issues.
even without BYOS (bring your own server) docs however, it's already possible to point TRMNL to your own stack if you 1) fork our OSS firmware + b) have some experience with e-ink.
Can you clarify what the difference between the Developer Edition and normal edition are? It's not clear from the checkout flow if this is required in order to create plugins, and is not mentioned anywhere in the docs.
They seem to have the api base url hardcoded in their firmware[1]. The repo seems to have pretty clear instructions for compiling and flashing modified firmware. From there, it's just a matter of writing a decent server to implement the calls documented in BYOD/S[2] and Private API.[3]
assuming your eink display would be on the same LAN as some always-on PC...
1. install python
2. make a file named `index.html` somewhere.
2a. put this in the "head" tag, so it'll refresh hourly: `<meta http-equiv="refresh" content="3600">`.
3. run `python -m http.server` from the same folder
This will start a single-threaded web server on 8000
4. On another machine on your network verify you can pull up http://firstmachine:8000/.
5. having proven it works, go buy an e-ink display and point it to http://firstmachine:8000/, make it the default homepage.
Voila.
Any time you have anything to say, just edit the `index.html` file and the eink display will update.
No need for fancy subscription services or kickstarter projects or crowdfunding... just... batteries included python.
Having done this, you will also most likely want to setup a javascript timer that also triggers a refresh in case the meta refresh fails. And a weekly reboot of the machine in case there is a memory leak or some other issue.
If you're comfortable with microcontrollers (esp32/arduino), I can definitely recommend Inkplate. I found them when I was making a similar setup for my parents, and they have various sizes up to 10" and up to 6 colors they can display.
You can either just get the module, or buy with a battery and mountable case already attached. I think all of the models are also available via Digikey and Mouser if people don't trust random websites.
You may be interested in https://github.com/aceinnolab/Inkycal, it looks like it's out of stock at the moment but they have pre-made devices or you can make your own with a list of parts.
My fridge isn't magnetic. A lot of modern fridges aren't.
Might be a neat idea to offer a magnetic mount for it, like a flexible flat magnetic board shaped to fit the TRMNL with a sticky backing so you can attach it somewhere and then use that to attach the TRMNL (your site doesn't seem to say anything about being magnetic so I'm guessing you have to attach magnets to the TRMNL too though?).
For that matter, the site doesn't offer any information about mounting it at all. Looking at the disassembly animation I see what looks like a hole to hang it on a nail, but it might be nice to put this info at least in the FAQ section if nowhere else (that does say it can be "hung on a wall" but no details).
If you have a hacker’s soul, an old Kindle, a jailbreak, and a Python installation, anything becomes possible. I’m working on something like that (though I hadn’t thought about markdown!). The Kindle is a particularly fun device once it’s hacked!
ah really?? feel free to email us (team@usetrmnl.com). we've been granted a bunch of EC licenses lately but maybe missed a few country check boxes on our store.
I have to deal with bouts with amnesia myself due to a dissociative disorder - this would be a really good appliance just to be able to see reminders to myself and also reminders from friends - relationships are really difficult to keep up with when its out of sight out of mind for anyone.
I'm working on a personal dashboard right now so I can have one space to leave notes for myself - I have the problem of not being able to consistently use the same tools since there are so many to reach out (social media, sms, chat apps, trello, physical notepads, .txt files). I frequently just fully forget that I've been taking notes every day, and where they're at. Building routines is, as one can imagine, really difficult. An app requires that I'm looking at my phone and can prioritize a notification. All the apps together are just too much to be able to prioritize, and I find myself hunting through all the apps for reminders or to try to ground myself.
I wish this had come up on HN (or I had had that idea myself) some years ago when my mother suffered from that same cruel condition, for the last four years of her life. With her body, all her older memories and her considerable intelligence largely intact, she had multiple moments of clarity every single day, in which she fully realized the terrible and hopeless situation she was in. But of course, within seconds this thought and any decisions she might have derived from it dissolved in the black hole of her defective short-term memory. So she would not even have had the ability to take her own life to end this if she wished so.
My brother and I tried many things to improve her life somewhat, only very few of those were actually a bit succesful. Two of them were digital gadgets, which we selected to provide some benefit without or at least with just very simple interactions: The best one was an LCD "picture frame" the only feature of which was to show an infinite loop of family photos stored on its SD card - she came to really like it and have it switched on quite consistently. The second one was an MP3 speaker which had a few hours of her favorite music on an SD card as well, and which could be used largely like a radio, just by pressing its play/stop button and volume buttons. This latter one she managed to enjoy at least from time to time.
Best wishes to the author and his mom, and everyone in a similar situation.
My wife acquired anterograde amnesia after a car accident. This device may or may not have worked for her: she would probably have discovered the device anew every time (as in, every 10 minutes or so), although she would probably be pleased each time.
Thankfully she fully recovered after a few weeks. It takes a lot of patience to deal with someone like that, and you could tell it frequently caused a lot of frustration on her part. Every 10 minutes or so in fact.
> One small challenge was maximizing the size of the message text. Sometimes a message is just a word or two; other times it might be several sentences. A single font size can’t accommodate such a wide range of text content. I couldn’t find a pure CSS way to automatically maximize font size so that a text element with word wrapping would display without clipping.
> I ended up writing a small JavaScript function to maximize font size: it makes the text invisible (via CSS visibility: hidden), tries displaying the text at a very large size, and then tries successively smaller font sizes until it finds a size that lets all the text fit. It then makes the text visible again.
Wow -- not just for accessibility but this seems like a very useful feature to have in native CSS.
Nice find.
Overall such a heartwarming use of technology. Love.
I've been watching the evolution of the web since 1995, and I remember when css got popular in the late 90s thinking that it didn't match real-world use cases. Somehow design-by-committee took us from drawing our sites with tables in the browser's WYSIWYG editor, to not being able to center text no matter how much frontend experience we have.
Css jumped the shark and today I'd vote to scrap it entirely, which I know is a strong and controversial statement. But I grew up with Microsoft Word and Aldus PageMaker, and desktop publishing was arguably better in the 1980s than it is today. Because everyone could use it to get real work done at their family-owned small businesses, long before we had the web or tech support. Why are we writing today's interfaces in what amounts to assembly language?
Anyway, I just discovered how float is really supposed to work with shape-outside. Here's an example that can be seen by clicking the Run code snippet button:
Notice how this tiny bit of markup flows like a magazine article. Browsers should have been able to do this from day one. But they were written by unix and PC people, not human interface experts like, say, Bill Atkinson. Just look at how many years it took outline fonts to work using strokes and shadows, so early websites couldn't even place text over images without looking like Myspace.
I think that css could benefit from knowing about the dimensions of container elements, sort of like with calc() and @media queries (although @media arguably shouldn't exist, because mobile shouldn't be its own thing either). And we should have more powerful typesetting metaphors than justify. Edit: that would adjust font size automatically to fit within a container element.
IMHO the original sin of css was that it tried to give everyone a cookie cutter media-agnostic layout tool, when we'd probably be better off with the more intuitive auto flow of Qt, dropping down to a constraint matrix like Apple's Auto Layout when needed.
Disclaimer: I'm a backend developer, and watching how much frontend effort is required to accomplish so little boggles my mind.
Your comment is some interesting food for thought, but I wanted to respond to a couple statements you made:
> not being able to center text no matter how much frontend experience we have
Not being able to center things is a bit of a meme, but flexbox was introduced back in 2009 and has been supported by major browsers for quite a long time. Centering text and elements is now extremely easy.
> css could benefit from knowing about the dimensions of container elements
You're in luck! Container queries were added to CSS fairly recently:
As someone who has struggled with getting CSS to do normal layout stuff that had clear precise semantics but required weird CSS trickery, it's actually more scary than lucky that stuff like container queries have arrived 30 years after CSS was introduced.
container queries have a very obvious chicken and egg problem if used a certain way: If this container is less than 30px wide, make its content 60px wide. Otherwise make it 20px wide. Now that container exists in a quantum state of being both 30 and 60px wide. I actually haven't looked into container queries to see how they ended up dealing with this yet.
Obviously this is a very contrived example but it can also express itself in subtler ways.
Really nice project. One idea for “if we fail to take down a message that no longer applies, it confuses her.” Put a start and end date/time on messages and implement in the board. That way you can pre schedule them and have them fall off automatically.
It might be nice to add default messages that can auto-populate the date so she won't notice if network goes down for awhile or someone forgets to post a message.
My dad didn't like poetry clock, but he does like image gen. So we got a (color) Inky Impression 7.3 and hooked it up to an RPi.
I made a basic telegram bot that you could send a verbal prompt to ("snowy day"). It would then ask which of your favorite artist styles it should create an image in. I found that presenting a list of two styles combined had cooler results. The prompt would be used to fetch a random quote on the topic, and quote and style would then be feed to stable diffusion, and maybe 30 seconds later you have fresh art and a quote on the display.
My dad then asked if we just could forward images directly there. He prefers, each day, to post an image of whatever the day is (November 13 is "World Kindness Day") and occasionally share a family photo. My mom looks forward to seeing what day he picks every day.
> There’s one other problem, though. It’s well known that AI language models like ChatGPT have a tendency to make up data (sometimes known as “hallucinations”), and it turns out that’s true even if you’re just telling the time. Roughly once every 15 minutes, says Webb, the clock will simply lie about the time just to make a certain rhyme work. “The fibbing is hilarious. Sometimes you can’t tell — it might say ‘one past two’ when it’s actually ‘two past one,’” he says. He says this will be fixable but, for now, is a fun quirk of the system. “Clockwork means you get precision drift; AI-work means you get hallucination drift.”
I love this. I wonder whether it would be useful to add a small indication of how roughly how long ago a message was written. For example, the first message might say 'this afternoon' or 'yesterday'. If it was written this afternoon, I know the dinner it's referring to is coming up, and wasn't yesterday.
Pimeroni has a selection of eink displays up to 7.3" including some with various buttons and LEDs to make whatever you'd like. https://shop.pimoroni.com/search?q=inky
All boox tablet/e-readers just run Android. They can do literally anything Android can for folks asking about the loading and displaying of the web page. There are several "kiosk" apps and browsers with kiosk modes. Also fairly expensive Android automation tools.
> It takes approximately 40 seconds to refresh this display
I think that would rule it out for the purpose of this project - the demo in their introduction video[0] shows that it flashes multiple colors for ages during this long refresh. I imagine that could be very confusing for someone whose short-term memory might not last that long.
Yeah, a black and white version would probably have been fine. Somewhat frustratingly they don't seem to have a 7 inch greyscale one though, only colour versions! Maybe they used to but stopped selling them?
EDIT: posted this before I noticed harpastrum's comment
If you're comfortable with microcontrollers (esp32/arduino), I can definitely recommend Inkplate. I found them when I was making a similar setup for my parents, and they have various sizes up to 10" and up to 6 colors they can display.
You can either just get the module, or buy with a battery and mountable case already attached. I think all of the models are also available via Digikey and Mouser if people don't trust random websites.
With an ESP32 or comparable (RPi Pico W, for example) you get MicroPython or CircuitPython support! That means a Python interpreter, drivers for popular peripherals and usually a network stack. Performance doesn’t beat a native SDK but Python is Python.
This is such a wonderful story, and I'm so happy that the author found something which worked well for their mom.
> Despite her amnesia, my mom came to remember that this display exits and what it’s for. She looks forward to seeing updates from her children on it.
This is the most interesting part for me here. Brains are such wondrous things. Would be cool to know if this is a special quirk of her mom or this is something which can help others like her too.
a commercial product along the same lines is KOMP https://komp.family/en/. We had it to communicate with our elderly grandparents until they died. Its a bit like a senior accessible social network feed for the family, including its dynamics, because the app shows what is being shown to everybody else of the family. In that regard it's a disadvantage you have some of the same dynamics going on. You dont only communicate to the grandparents, but also to the (extended) family.
I don't know if it's because we are conditioned by our interaction with TV and mobiles, but active LCD screens feel like they are screaming for our attention and an always on display will mostly be a distraction.
E-ink displays don't have this, they just blend in.
Understanding that the condition is rare enough that most of us really don't have a need to prepare for it, I wonder if there are any habits one could cultivate that would make it easier to live with amnesia. Learning new things is my favorite past time and strongest coping mechanism, so the though of not being able to do that anymore is up there with locked-in syndrome on my list of greatest living fears.
For example, I am already in the habit of logging every phone call to any doctor's offices or important contacts as they're happening. Being able to refer back to all the notes has helped me manage a number of complex errors. I know the name of the person I spoke to, the date, and what we discussed. Any time I need to make a call about a topic or to a company, I have an easy way to pull up all the past notes.
I'd like to think if I ever got amnesia, already having this system in place would serve me really well if I couldn't learn new things. I have the old things, and the habit of referring to and adding new things to the list.
But I wonder what else would or wouldn't be useful to try to practice now?
If I'm right that this condition is like that of Henry Molaison - https://en.wikipedia.org/wiki/Henry_Molaison - then the real difficulty is that you don't remember that you have amnesia.
Well yes, but my current "write down the details of my calls and refer back to them every time" wouldn't require me to remember I had amnesia, right? For now, I do remember those past conversations, but if I stopped remembering them, having them up on my screen in the side panel of my note-taking app would still make them available to me if I didn't.
Sign language and brail come to mind as useful in this regard (if not for you, then for a loved one).
As for amnesia, it seems like a habit of making notes and seeking out to read your own notes would be useful. However, the trend in technology to constantly change behaviour, appearance, and functionality makes anything digital a barrier. Manual notes are also susceptible to being impossible for ageing people to make. So it's really hard to think of something.
I have visited the thought of what it’d be like to have amnesia like this many times throughout my entire life. I am sure reality is nothing like my thoughts, but in fantasy land it’s just interesting to imagine picking up a note in my own hand writing saying “you have amnesia, everything is okay, everyone is well and happy, some bathroom humor, go watch YouTube and chill”
Does anyone know if "Start its web browser and have that browser display a designated start page." is specific thing for this tablet or if that is "normal" in android?
I want to do something similar for anki cards I'm struggling with, and I dunno if I'm in for a world of pain. I was considering https://shop.boox.com/products/go6 for my needs as it's a bit cheaper.
I love these kinds of projects. Congratulations to the OP.
Unrelated, but does anyone know a good TV remote for elders? I'd like something like a Stream Deck with big buttons for things like :
* Turn it on/off
* Switch to TV channel 315
* Switch to TV channel 517
* Play Planet Earth on Netflix
* Play Young Sheldon on Netflix
My grandparents are 92 and 97 and even big remotes aren't cutting it. Not only that, but I'd like for them to be able to use ondemand video platforms, not only random TV channels.
To control the TV itself, it seems a RPi or ESP32 with an IR led is enough, but to put something to play on Netflix is surprisingly difficult. I'm able to control a Fire Stick using remote adb commands, but not sure how reliable it is. I'd love to find something like this off the shelf.
Technology is great, but it's not made for elders. It frustrates them (and me), and they end up feeling stupid, which angers me.
I am sure someone else must have done this, but I couldn't find it anywhere.
Avoid toggling buttons. Have a separate button for on, and a second button for off. Idempotent all the way.
I remember my great aunt repeatedly mashing the on/off button insisting that the TV was not working, when it never had a chance to bring up the picture.
Is their problem the size (big buttons) or the UI complexity the buttons control (trying to navigate on Netflix is a PITA compared to hitting a preset channel button)?
On the former I had luck with one of those jumbo remotes that just has a few buttons (channel up/down, volume up/down, power, mute) and separately programming the TV to only have the channels they cared about in the list. When it came to smart apps it just became impossible to try to fix via the remote as the remote wasn't really the problem.
The author mentions needing a small storage service and paying for JSONStorage. In fact, when I've needed this sort of thing in the past, Google Sheets is an amazing tiny database for projects. It has pretty generous API rate limits and it's convenient to be able to manage your DB live in a spreadsheet. In fact, even levels.fyi got away with a Google Forms / Google Sheets backend for a long time[1].
>Since the physical device was satisfactory, the next step was writing a simple website that could drive the display.
>A Compose page my siblings and I write messages and save them to be displayed.
Is there a risk of a malicious actor discovering the website and writing in their own messages? I would think building user authentication in to the MomBoard website might be a bit heavyweight. Whats the best way to do this?
When you have a site with a fixed, tiny amount of users, I'd opt for HTTP basic auth (via HTTPS). Whether you're using nginx, Traefik, Caddy, etc..., it's very easy to setup. If you're using something like Cloudflare Pages, I would guess you could setup a worker to handle it for you (though I'm not familiar enough with workers to be sure).
Love this! A couple years ago I had to do similar for my grandma (94 at the time) who was losing her hearing as well as short term memory loss. Was surprised how few off the shelf options there were.
Had to write a medium article with exact instructions for how to use it otherwise she was very skeptical of using it. The only issue was that she would sometimes read aloud which would enter her into a feedback loop lol.
My sister has a disability making independent living a challenge. Although I have 0 technical background, I need to start thinking and brainstorming in this manner.
> I ended up writing a small JavaScript function to maximize font size: it makes the text invisible (via CSS visibility: hidden), tries displaying the text at a very large size, and then tries successively smaller font sizes until it finds a size that lets all the text fit. It then makes the text visible again.
That would be a good application for dichotomic search if performance was ever a problem (I doubt it though).
More generally, having elements on a grid of different sizes should hopefully be much more easy once CSS masonry grid is available.
This is awesome and I am happy to read that she was able to remember the device and asks if things have been added to it. My parents have just retired and I wonder if something like this would be advantageous to introduce prior to signs of memory loss. My grandmother had Alzheimers and while it is different than the amnesia that OP references, her memories were lost in reverse chronological order (can't remember where her keys are, but could remember where her last job was, later could not remember that last job, but could remember her first job, etc). So introducing this prior to those recent memory lapses could help solidify that device in my parents head so that they could benefit from it even if they do start to exhibit that behavior.
Does any commercial version of this exist? Using an existing tablet makes the DIY aspect a little less but then you have to roll your own site as well.
It might be worthwhile to look at how an LLM might assist someone with this condition. A lot of the persistence hacks used on LLM chatbots are addressing the lack of retained memories outside of the context window, so maybe something like RAG could help your mom live a less limited life, or reduce some of the burden on you or other caretakers.
Brilliant use of tech, I'm happy when I see someone turn their nerd-powers to things that unabashedly make life better. Good work!
This is a very interesting idea. My father has dementia. I'm not sure this would really work in this use case. He wouldn't remember to look at the display. Like I said, I'm not sure about this. It might be worth trying though.
I have always dreamt of something like this to connect with aging loved ones.
I would love something that they could also listen to with a big button to play recent messages.
A little off topic, but on the subject of E-Ink, here is an analysis of a Kindle display with optical coherence tomography images: https://arxiv.org/abs/1605.05174
Glad I went on HN today. My grandma has dementia and I've been leaving paper reminders around the house. Maybe I should try something like this. Wishing you and your family the best.
Interesting. Waveshare makes this recommendation across their product line, and from what I can glean from google, their displays really do develop permanent artifacts if they're not refreshed. I've never seen another epaper company say this, and anecdotally I've seen ereaders that held the same image for months or years work perfectly fine, even with no power for refreshes. I don't know why there would be a difference.
I encouraged my mother, who had short-term memory issues _and_ dementia to write things down. It backfired. She's write something down, like the fact that she got a call from her gardener, then obsess over it. She'd open the book over and over and then talk about the "odd" call she got from her gardener for weeks! We had to take the book and write things like "This issue has been resolved" under the things she wrote down.
I think an e-ink that we control remotely might work for her, too. We can put an item up there and then remove it as soon as it's not relevant anymore so she won't keep re-reading it and obsessing over it.
This is the kind of feedback I was looking for. These are not conditions that we are familiar with - and so the user interface for such things is not necesarily obvious. For example:
- I would expect that their and your mom did not forget how to use a web browser when they acquired short term memory issues. So a web browser interface (links, scrolling, buttons) might still be fine.
- I was surprised that in the posters' display, the date / time of each msg was not shown.
- I was susprised that there is no reminder "Mum, since an operation, you rarely form short term memories." Wouldn't it make sense that even that is not forming a memory?
- As you mention, remove the item or mention how it was resolved.
When I watched the film Memento, I found myself thinking 'holy shit, I'm not quite far away enough from this...'. No tattoos yet, but one could write a book on the stcky pads I've laying around.
As a bit of a luddite, e-ink is one of the few modern wizbangs I'm enamored with. It's so damn nice I take it as an inside woosh joke that it isn't everywhere and available without pawning my organs.
(not the original poster), a reminder that changing the law is important in many situations and can change real-life impact of people (for example with this tablet, or with blood oxygen feature of Apple) ?
If we had truly enforced software patents we wouldn't see widespread LLMs.
As a side note on his mother remembering that the tablet exists, it sounds like she has amnesia quite like Henry Molaison, a famous case study in neuropathology. He had very specific brain damage that seemingly stopped him forming new memories in the same way as OP's mother, but studies showed that he could remember some things, just not consciously. So for example he would have warm feelings towards people who'd been caring for him despite not remembering them, and would also pick up card games more and more quickly as he played them repeatedly despite saying he didn't remember the game. OP's mother remembering the tablet sounds very similar, particularly when paired with the feeling of being remembered and loved by her children.
This reminds me of muscle memory. I can play pieces on the piano even though I don't actively remember the sheet music of them. My hands just "know" what to do. Funnily enough the moment I start actively thinking about certain passages that ability worsens by a lot.
At the start it's all about carrying around notes full of picking the relevant condition depending on the current permutation/state of the cube then following the step by step algorithms on which sequence of steps to perform for that condition.
Then you'll naturally realise that certain conditions happen a lot more than others and you'll start to remember the sequence of letters for each series of steps to perform.
Over time you'll forget the letters and your fingers will just know the sequence to perform when you perceive that condition, kind of like typing a password without thinking about it.
Eventually you'll be able to fit each condition and algorithm into your muscle memory and completely forget the series of letters that you used to memorise.
Now I can barely explain how to solve a rubik's cube in-person. I just do it.
I think that's when you really know a piece, when you can play it incredibly slowly. Paradoxically it's easy to play quickly and just let your fingers play out their muscle memory, playing something really slowly is the challenge.
Now I’m wondering if you can tell a kid is from an “even” or “odd” generation by which way they tie shoes…
https://www.sciencedirect.com/science/article/abs/pii/S03069...
Reddit is telling me to not accept it at face value - https://old.reddit.com/r/research/comments/1bh2jmv/this_is_h...
These mechanisms may or may not encode memories as we typically understand them, i.e., the ability to remember an event or fact, but could very plausibly shift personality, preferences, etc.
> organs outside of the brain don't contribute to memory
Interesting question. To start, personality typically refers to the totality of a person's behaviors, not the memories they may be able to bring forth. Behavior, esp automatic, is informed by cognitive states informed by the body.
Affect is the general sense of feeling that you experience throughout each day. It is not emotion but a much simpler feeling with two features. The first is how pleasant or unpleasant you feel, which scientists call valence. . . . The second feature of affect is how calm or agitated you feel, which is called arousal. [0]
Simple pleasant and unpleasant feelings come from an ongoing process inside you called interoception. Interoception is your brain’s representation of all sensations from your internal organs and tissues, the hormones in your blood, and your immune system.
...[M]oment-to-moment interoception infuses us with affect, which we then use as evidence about the world. People like to say that seeing is believing, but affective realism demonstrates that believing is seeing.
0. Barrett, Lisa Feldman. How Emotions Are Made: The Secret Life of the Brain (p. 72). HarperCollins. Kindle Edition.
1. ibid (p. 56).
2. ibid (pp. 76-77)
At least playing is mostly an entertainment. Passwords is where the shit happens. I recently lost a 20y old account thanks to this.
I expect it is very hard to overestimate how incorrect our mental model memory and learning is. If literally everything was forgotten, then you could set up a reverse groundhog day or groundhog hour for someone, just optimize for them having a wonderful day every day. (Would still be horrible for the loved ones to be effectively disconnected from their still-living relative.) Probably there have been movies made about this.
I have no experience with this but I am sure it is nothing, nothing, nothing like that. The article says you wouldn't wish it on your worst enemy.
> Because she cannot remember things, she goes through each day in a state of low-grade anxiety about where her grown children are and whether they are all right. She feels she hasn’t heard from any of us in a long time.
To me this is not a description of someone frozen in time. To me this is a description of some horrific combination of some amount of learning or "remembering" happening, some sense of passage of time, and no episodic memories to draw on to explain any of it.
There is a Drew Barrymore movie Fifty first dates. And yes, it is horrible for the relatives.
If nothing else, myelinization counts as a form of memory. Strengthened by reuse.
I would love to know if those warm feelings are stronger with individuals who remind you of someone you used to know. “This nurse reminds me of Aunt Sarah, who was nice to me when my dog died.” And so forth.
I have no idea if feelings would automatically transfer to me from people with amnesia, but they certainly do for people without it, even though I don't remind them of anyone they know well enough to name.
I always find those non-obvious connections fascinating, like the disorders where e.g. someone can't say the word "fork" when they're looking at one despite being to describe what you use it for etc, but can immediately name it when they touch it.
Edit: got a link? I'd be interested to read that.
I have a relative with anterograde amnesia from a stroke, so that story got passed on to my father when it happened. 8 years ago perhaps?
It's even aesthetically pleasing! What mom wouldn't find this charming?
This reminds me of a desire I've had for a long time: a simple, wall-mountable eInk device that could be configured with a URL (+wifi creds) and render a markdown file, refreshing once every hour or so. It would be so useful for so many applications – I'm a parish priest and so I could use it to let people know what events are on, if a service is cancelled, the current prayer list, ... the applications would be endless. I'd definitely pay a couple of hundred dollars per device for a solid version of such a thing, if it could be mounted and then recharged every month or two.
I wanted the same kind of general eink device, but this is also supposedly super hackable!
no longer a Kickstarter btw, shipping same-day now (see homepage)
(This is fantastic. Thank you for sharing about it!)
> Your TRMNL device pings our server, never the other way around.
> Each request made to our /api/display endpoint includes only the minimum details needed to support customers -- an API key, device mac address, firmware version, battery voltage, and wifi signal strength.
Super hackable but it pings their hosted server and nothing else?! Is there a way to run your own server?
if you think about it, we are incentivized to do this. no subscription fees means the more you ping our server, the lower our margin. but for now we're wrapping up fulfilling all pre-orders, scaling, etc typical new product issues.
even without BYOS (bring your own server) docs however, it's already possible to point TRMNL to your own stack if you 1) fork our OSS firmware + b) have some experience with e-ink.
> Purchase a TRMNL from our home page: https://usetrmnl.com
> Then follow the instructions on BYOD/S > Server.
> More TBD.
[1]: https://github.com/usetrmnl/firmware/blob/e3db8c37990c2333ec...
[2]: https://docs.usetrmnl.com/go/diy/byod-s
[3]: https://docs.usetrmnl.com/go/private-api/introduction
Any time you have anything to say, just edit the `index.html` file and the eink display will update.
No need for fancy subscription services or kickstarter projects or crowdfunding... just... batteries included python.
You can either just get the module, or buy with a battery and mountable case already attached. I think all of the models are also available via Digikey and Mouser if people don't trust random websites.
https://soldered.com/categories/inkplate/
(disclaimer, i'm the founder)
Might be a neat idea to offer a magnetic mount for it, like a flexible flat magnetic board shaped to fit the TRMNL with a sticky backing so you can attach it somewhere and then use that to attach the TRMNL (your site doesn't seem to say anything about being magnetic so I'm guessing you have to attach magnets to the TRMNL too though?).
For that matter, the site doesn't offer any information about mounting it at all. Looking at the disassembly animation I see what looks like a hole to hang it on a nail, but it might be nice to put this info at least in the FAQ section if nowhere else (that does say it can be "hung on a wall" but no details).
https://usetrmnl.com/assets/section2-3-d6887b41db12ad0659992...
as the first character, タ (ta), is missing from the display, making it read "(a)minaru".
Also, what country are the orders shipped from? US?
It's shipped from USA.
If eInk wasn't a monopoly this would be 100% a project I'd love to do
I'm working on a personal dashboard right now so I can have one space to leave notes for myself - I have the problem of not being able to consistently use the same tools since there are so many to reach out (social media, sms, chat apps, trello, physical notepads, .txt files). I frequently just fully forget that I've been taking notes every day, and where they're at. Building routines is, as one can imagine, really difficult. An app requires that I'm looking at my phone and can prioritize a notification. All the apps together are just too much to be able to prioritize, and I find myself hunting through all the apps for reminders or to try to ground myself.
Thankfully she fully recovered after a few weeks. It takes a lot of patience to deal with someone like that, and you could tell it frequently caused a lot of frustration on her part. Every 10 minutes or so in fact.
> I ended up writing a small JavaScript function to maximize font size: it makes the text invisible (via CSS visibility: hidden), tries displaying the text at a very large size, and then tries successively smaller font sizes until it finds a size that lets all the text fit. It then makes the text visible again.
Wow -- not just for accessibility but this seems like a very useful feature to have in native CSS.
Nice find.
Overall such a heartwarming use of technology. Love.
Css jumped the shark and today I'd vote to scrap it entirely, which I know is a strong and controversial statement. But I grew up with Microsoft Word and Aldus PageMaker, and desktop publishing was arguably better in the 1980s than it is today. Because everyone could use it to get real work done at their family-owned small businesses, long before we had the web or tech support. Why are we writing today's interfaces in what amounts to assembly language?
Anyway, I just discovered how float is really supposed to work with shape-outside. Here's an example that can be seen by clicking the Run code snippet button:
https://stackoverflow.com/a/33953666
Notice how this tiny bit of markup flows like a magazine article. Browsers should have been able to do this from day one. But they were written by unix and PC people, not human interface experts like, say, Bill Atkinson. Just look at how many years it took outline fonts to work using strokes and shadows, so early websites couldn't even place text over images without looking like Myspace.
I think that css could benefit from knowing about the dimensions of container elements, sort of like with calc() and @media queries (although @media arguably shouldn't exist, because mobile shouldn't be its own thing either). And we should have more powerful typesetting metaphors than justify. Edit: that would adjust font size automatically to fit within a container element.
IMHO the original sin of css was that it tried to give everyone a cookie cutter media-agnostic layout tool, when we'd probably be better off with the more intuitive auto flow of Qt, dropping down to a constraint matrix like Apple's Auto Layout when needed.
Disclaimer: I'm a backend developer, and watching how much frontend effort is required to accomplish so little boggles my mind.
> not being able to center text no matter how much frontend experience we have
Not being able to center things is a bit of a meme, but flexbox was introduced back in 2009 and has been supported by major browsers for quite a long time. Centering text and elements is now extremely easy.
> css could benefit from knowing about the dimensions of container elements
You're in luck! Container queries were added to CSS fairly recently:
https://developer.mozilla.org/en-US/docs/Web/CSS/CSS_contain...
I agree with GP that CSS should be scrapped.
Obviously this is a very contrived example but it can also express itself in subtler ways.
My dad didn't like poetry clock, but he does like image gen. So we got a (color) Inky Impression 7.3 and hooked it up to an RPi.
I made a basic telegram bot that you could send a verbal prompt to ("snowy day"). It would then ask which of your favorite artist styles it should create an image in. I found that presenting a list of two styles combined had cooler results. The prompt would be used to fetch a random quote on the topic, and quote and style would then be feed to stable diffusion, and maybe 30 seconds later you have fresh art and a quote on the display.
My dad then asked if we just could forward images directly there. He prefers, each day, to post an image of whatever the day is (November 13 is "World Kindness Day") and occasionally share a family photo. My mom looks forward to seeing what day he picks every day.
That's fun. Although, from the article:
> There’s one other problem, though. It’s well known that AI language models like ChatGPT have a tendency to make up data (sometimes known as “hallucinations”), and it turns out that’s true even if you’re just telling the time. Roughly once every 15 minutes, says Webb, the clock will simply lie about the time just to make a certain rhyme work. “The fibbing is hilarious. Sometimes you can’t tell — it might say ‘one past two’ when it’s actually ‘two past one,’” he says. He says this will be fixable but, for now, is a fun quirk of the system. “Clockwork means you get precision drift; AI-work means you get hallucination drift.”
;)
All boox tablet/e-readers just run Android. They can do literally anything Android can for folks asking about the loading and displaying of the web page. There are several "kiosk" apps and browsers with kiosk modes. Also fairly expensive Android automation tools.
> It takes approximately 40 seconds to refresh this display
I think that would rule it out for the purpose of this project - the demo in their introduction video[0] shows that it flashes multiple colors for ages during this long refresh. I imagine that could be very confusing for someone whose short-term memory might not last that long.
[0] https://www.youtube.com/watch?v=TluopgSoSWY&t=500s
EDIT: posted this before I noticed harpastrum's comment
You can either just get the module, or buy with a battery and mountable case already attached. I think all of the models are also available via Digikey and Mouser if people don't trust random websites.
https://soldered.com/categories/inkplate/
> Despite her amnesia, my mom came to remember that this display exits and what it’s for. She looks forward to seeing updates from her children on it.
This is the most interesting part for me here. Brains are such wondrous things. Would be cool to know if this is a special quirk of her mom or this is something which can help others like her too.
[1] https://en.wikipedia.org/wiki/Henry_Molaison [2] https://en.wikipedia.org/wiki/Repetition_priming
E-ink displays don't have this, they just blend in.
For example, I am already in the habit of logging every phone call to any doctor's offices or important contacts as they're happening. Being able to refer back to all the notes has helped me manage a number of complex errors. I know the name of the person I spoke to, the date, and what we discussed. Any time I need to make a call about a topic or to a company, I have an easy way to pull up all the past notes.
I'd like to think if I ever got amnesia, already having this system in place would serve me really well if I couldn't learn new things. I have the old things, and the habit of referring to and adding new things to the list.
But I wonder what else would or wouldn't be useful to try to practice now?
That's the kind of idea I'm looking for.
As for amnesia, it seems like a habit of making notes and seeking out to read your own notes would be useful. However, the trend in technology to constantly change behaviour, appearance, and functionality makes anything digital a barrier. Manual notes are also susceptible to being impossible for ageing people to make. So it's really hard to think of something.
I want to do something similar for anki cards I'm struggling with, and I dunno if I'm in for a world of pain. I was considering https://shop.boox.com/products/go6 for my needs as it's a bit cheaper.
Unrelated, but does anyone know a good TV remote for elders? I'd like something like a Stream Deck with big buttons for things like :
* Turn it on/off
* Switch to TV channel 315
* Switch to TV channel 517
* Play Planet Earth on Netflix
* Play Young Sheldon on Netflix
My grandparents are 92 and 97 and even big remotes aren't cutting it. Not only that, but I'd like for them to be able to use ondemand video platforms, not only random TV channels.
To control the TV itself, it seems a RPi or ESP32 with an IR led is enough, but to put something to play on Netflix is surprisingly difficult. I'm able to control a Fire Stick using remote adb commands, but not sure how reliable it is. I'd love to find something like this off the shelf.
Technology is great, but it's not made for elders. It frustrates them (and me), and they end up feeling stupid, which angers me.
I am sure someone else must have done this, but I couldn't find it anywhere.
I remember my great aunt repeatedly mashing the on/off button insisting that the TV was not working, when it never had a chance to bring up the picture.
On the former I had luck with one of those jumbo remotes that just has a few buttons (channel up/down, volume up/down, power, mute) and separately programming the TV to only have the channels they cared about in the list. When it came to smart apps it just became impossible to try to fix via the remote as the remote wasn't really the problem.
If anyone has a solution in this space, I would be very interested.
1. https://www.levels.fyi/blog/scaling-to-millions-with-google-...
>A Compose page my siblings and I write messages and save them to be displayed.
Is there a risk of a malicious actor discovering the website and writing in their own messages? I would think building user authentication in to the MomBoard website might be a bit heavyweight. Whats the best way to do this?
https://developers.cloudflare.com/workers/examples/basic-aut...
Had to write a medium article with exact instructions for how to use it otherwise she was very skeptical of using it. The only issue was that she would sometimes read aloud which would enter her into a feedback loop lol.
https://medium.com/@admangan2018/how-to-utilize-the-transcri...
My sister has a disability making independent living a challenge. Although I have 0 technical background, I need to start thinking and brainstorming in this manner.
Meeting for dinner tonight? Set the message to expire after today.
> I ended up writing a small JavaScript function to maximize font size: it makes the text invisible (via CSS visibility: hidden), tries displaying the text at a very large size, and then tries successively smaller font sizes until it finds a size that lets all the text fit. It then makes the text visible again.
That would be a good application for dichotomic search if performance was ever a problem (I doubt it though).
More generally, having elements on a grid of different sizes should hopefully be much more easy once CSS masonry grid is available.
Disclosure: I'm from the TRMNL team.
[1] - https://github.com/usetrmnl/plugins
The creator is on HN too.
Brilliant use of tech, I'm happy when I see someone turn their nerd-powers to things that unabashedly make life better. Good work!
It solved a personal problem.
Amazing mission behind the tech.
Could solve a myriad of issues for other families. This part is unproven, but that's why it would be cool to see the author release it as a product!
Could start by simply putting up a payment page and making them bespoke as orders start coming in.
Could tweak the UI to add an expiration date on the initial input screen, with a sensible default (maybe 2 weeks?)
Happily, e-ink displays don't suffer from burn-in.
Can't say if it is due to burn-in but some manufacturers do recommend refreshing the display image periodically like every 24 hours [1].
See precautions #4
https://www.waveshare.com/wiki/4.2inch_e-Paper_Module_(B)_Ma...
I think an e-ink that we control remotely might work for her, too. We can put an item up there and then remove it as soon as it's not relevant anymore so she won't keep re-reading it and obsessing over it.
- I would expect that their and your mom did not forget how to use a web browser when they acquired short term memory issues. So a web browser interface (links, scrolling, buttons) might still be fine.
- I was surprised that in the posters' display, the date / time of each msg was not shown.
- I was susprised that there is no reminder "Mum, since an operation, you rarely form short term memories." Wouldn't it make sense that even that is not forming a memory?
- As you mention, remove the item or mention how it was resolved.
As a bit of a luddite, e-ink is one of the few modern wizbangs I'm enamored with. It's so damn nice I take it as an inside woosh joke that it isn't everywhere and available without pawning my organs.
If we had truly enforced software patents we wouldn't see widespread LLMs.