Some Thoughts on Smart Canes | Hacker News Reader

Some Thoughts on Smart Canes

derekriemer.com

101 pointsgoranmoomin62 comments

Pretty solid arguments. I had an idea once for an addition to a cane specifically for blind people trying to learn or use echolocation. This would be a periodic noisemaker built into the cane so the blind person wouldn't have to chirp themselves.

I guess there's no reason why this would have to be in a cane as opposed to a little device you could store in our pocket or on your belt. When I was imagining it I thought the cane would be a good spot to put a battery plus device, but seeing so much emphasis on weight from this article - maybe not.

Did you think of this idea before smartphones were invented?

There are plenty of electronic devices that are still valuable as a discrete object instead of an app that requires navigating your smartphone's ui to use (obviously).

The article gives one specific reason: "I've walked several miles in weather cold enough to become frost bitten in minutes, and I can't aford to take my gloves off to operate a touch screen."

The article mentions GPS specifically as not an important part of a smart cane. Other features need to be easier to access, or more dependable than a phone which might die because of all the other things draining its battery.

I'm only imagining this tool, I've done no research on it. In my imagination you may need to point the sound, you may need specialized speakers to get the right frequency and duration for the click, and you might not want the clicker to be in your pocket / might not want to be holding your phone out of your pocket for long periods of time.

I had the fortune of working for a very small company that made electronic vision enhancers and portable readers for visually impaired people. We were just 2 developers and there were about 15 people in total mostly doing electronics, assembly and marketing.

It was very surprising how dumb our intuitions can be about what's important for our customers and what isn't. And how easy it is in daily grind to make something that can't actually be used by a blind or elderly person. I think they would be better of if they hired at least one blind developer, even if (s)he worked a little slower.

Why would you assume they would work slower?

I don't know if they would. I said it would be worth it even if they did.

Would you add that caveat for any other demographic?

If I wrote they should be hired from a significantly reduced pool of candidates - sure.

> "Why would you assume they would work slower?"

I'm legally blind and I work slower, in some regards, particularly when working with text produced by other people. Most of the colleagues I knew who are legally blind also had similar limitations. The other poster's assumption is not unjustified.

As someone who’s lost most of my proprioceptic sense of balance[0], I use a cane for two things. The main one is as replacement sense of balance; having the cane resting even very lightly on the ground fills in some gaps left by my feet and gives me a feel for my orientation. The cane is also helpful in minimizing risk from falls. That’s where I think there’s room for some improvements. If my had a way of detecting either that I was tipping over, or (easier), that I’ve hit the ground very hard with the tip, and were able to deploy some additional stabilization, that would be really helpful.

[0] https://en.wikipedia.org/wiki/Proprioception

I've seen some studies saying that hats with gyroscopes are effective, but they focused almost entirely on an exotic output device that worked via sensory substitution. What am I missing about "hat with gyroscopes and tiny servos that tap you on the side of the head in the direction that's down"? I assume that's the first thing that everyone would try but I haven't heard about it being a common medical device, so there must be something wrong with it.

Or is it the "deploy some additional stabilization" step that's missing? Can't think of a realistic way to do that, true. Anything solid enough and deployed quickly enough to be useful would be dangerous. Can't just beef up the cane, you'd have already tried that if it was useful.

Could be as simple as spring-loaded extra feet that lock into place.

Not visually impaired but I have used canes for several stretches over the past few years while healing from athletic injuries. For that use case the one innovation was the collapsible cane which was handy for public transport, planes, and driving (only time I appreciated,nor used, automatic transmission). I always felt like a “tourist” using a cane.

I do use hiking poles and though my camping preference is to buy as absolutely little specialized gear as possible, I was glad when I switched from ski poles to adjustable hiking poles.

It’s all about the domain, as this article so nicely points out.

While we're at it, can we stop replacing physical knobs and buttons by dumb touch sensitive buttons ? I feel like a cave man when I have to repeatedly press "+" on my induction hob and hear the stupid beep it makes x times. Plus it does not work well when it's wet which of course happens a lot, it's a kitchen ffs. A good example of "backwards" progress unless of course the ease of cleaning your hob is more important to you than using it for you know, cooking.

Using my induction hob is a nightmare for this reason.

Another possible reason for not having physical buttons is waterproofing. There would need to be a small gap between the button and the hob surface, and water or cooking oil could get in there, causing issues with the electronics, in addition to making it harder to clean.

How common was water damage to hobs before physical buttons were removed? I'm guessing not very. Maybe there's a bit of cost saving involved when manufactoring hobs without buttons. But I'd much rather pay a couple of extra currency units and get something that's user friendly.

> How common was water damage to hobs before physical buttons were removed? I'm guessing not very.

I think this is correct - not much damage. I used gas hobs for decades with various types of buttons and knobs. With appropriate recesses, washers, etc not much spilt water got past them, but the resultant complex surfaces were difficult to clean.

I liked hardware buttons hidden beneath a polymer layer (to simulate the look of touch buttons). Easy to use, clean and waterproof. The downside was that the plastic layer sometimes came off after a few years on cheap appliances and needed re-gluing. Better than a failed touch panel, though.

This is actually a very easy problem to solve with minimal design : Just find a way to isolate the knobs from the main cooking area. Like this but with induction : https://duckduckgo.com/?q=gas+hob+&t=fpas&iax=images&ia=imag...

I find that they're not even easier to clean. When I clean my hob and the button area gets a bit wet, it will randomly turn on. Turning cleaning into a potentially painful affair.

Also currently have an induction hob and hate it for the very same reason. Well, I also hate it because induction is complete crap for cooking. Mine has two settings: 0-6 = nothing, 7 = burnt. I've no idea what 8 and 9 are for (maybe boiling water, but it's still slower than my kettle). Then add the fact that it tries to regulate heat by cycling on and off, meaning you cycle between burning and nothing, and it makes an annoying sound while doing so. Plus the fact it heats a pathetically small area in the middle of the pan and nothing else, making frying an egg way more difficult than it should be.

Luckily I'm going back to gas very soon.

I assume you already have experimented with different pans, and have experienced the same difficulties with all of them?

I would agree that gas is still superior, but overall, induction has been working extremely well for us, and the safety benefit around smaller children may outweigh the culinary benefits of gas cooking for us.

Gas is also toxic.

Around here, the gas used for cooking is natural gas, which is not toxic in the technical sense of the word, though one could theoretically suffocate from it.

I say "theoretically" because I can't recall hearing of a single instance of such a death, while deaths from e.g. carbon monoxide poisoning are far from uncommon.

It feels toxic enough. I always open the back door when we have our natural gas oven on because it makes breathing unpleasant otherwise.

That is the additional gas that is mixed in to help you notice a leak. Natural gas itself is odourless and you’re unlikely to notice it.

Natural gas stoves still have health risks:

https://www.vox.com/energy-and-environment/2020/5/7/21247602...

Interesting! I did not know much of that.

I have high quality cookware, mostly stainless steel and cast iron. They're not the same brand or anything. I'm not prepared to spend hundreds of pounds experimenting with different types of pans. I highly doubt it would make a difference anyway. Can't wait to be back on gas.

Yeah, I did not mean to suggest you buy different pans, merely that you try a few of those you already have. But it sounds like you just have a poorly performing induction range, where no pan in the world can help.

This is not, however, inherent in the technology as such.

> While we're at it, can we stop replacing physical knobs and buttons by dumb touch sensitive buttons ? I feel like a cave man when I have to repeatedly press "+" on my induction hob and hear the stupid beep it makes x times.

Most induction systems can be controlled by ovens that have physical knobs. For example the AEG EU327MDE [1]. The hob-oven interface is more or less standardized. For unfathomable reasons such combined "induction hobs + compatible oven with physical controls" are not sold in all markets worldwide.

[1] https://www.euronics.de/haus-und-haushalt/kochen-und-backen/...

There's a place for everything. My cpap machine has the loudest clickiest physical button of anything in my house, much to my wife's displeasure. I wonder every day why they put a heavy clicky button on something you use in bed with a sleeping partner.

So next year they can sell "quieter buttons" as a new feature?

Perhaps so you can tell if it activated or not. Plus their user base skews older so some proportion may have hearing problems.

I don’t use a CPAP machine so don’t know if this matters (perhaps it’s obvious when it’s on/off). Just speculating on your “I wonder every day” comment.

Also the CPAP machine is regulated as a medical device, so then fr is required by the FDA to certify that the machine has a certain lifespan, which includes cycle times on things like switches. It’s likely that a big mechanical device which is over specified in its lifetime is an easy way to solve a problem so design attention can be directed elsewhere.

Even so, I can't imagine there isn't a less clucky button.

It's very obvious when it's on or off. The machine has been well designed with a11y in mind, so it lights up and has icons. Also, it inflates your upper respiratory tract like a balloon, which is pretty hard to miss.

My Kia Niro kept physical buttons, but uses up/down buttons for fan speed. It drives me crazy. I wish ranges were knobs and buttons were discrete selectors. Touch controls are distractors while driving.

> A good example of "backwards" progress unless of course the ease of cleaning your hob is more important to you than using it for you know, cooking.

They’re also fewer parts, more resistant to dirt, can’t break off, takes less space during shipping, and looks better in brochures and showrooms.

I’m not saying I prefer using them but I am not surprised at their success.

Another reason is apparently modularity. There’s an increasing demand in wider stovetops with more or larger heaters, induction and so on, and having all controls on the top itself means users can combine all ovens with all stovetops. Additionally, ovens at eye-height are becoming more popular, so in some cases there isn’t even an oven below the cooktop.

You could have physical knobs on the top, but those interfere with the pots and sealing them against spills is hard. The only decent solutions I’ve seen are magnetic knobs and a dedicated knob bar that is installed at the face, with drawers below it, but those are much more expensive to produce.

That's a UI design choice not inherent to the functionality. My induction stove has a touch strip where I can choose the heat intensity directly without having to go through intermediate states. It's actual more convenient than the analog dial on my gas range, in a sense, because the gas range dial only has low/medium/high zones, and I have to remember 2/3 medium for eggs and 1/4 high for steak, as opposed to 3 for eggs and 8 for steak on the induction stove.

I don't know why but my brain kept inserting an R into cane. It took a while for my brain to unsee the R.

I am all for equality and enablement but I was downright terrified of crane operators with visual impairments. The imagery of such an operator on a dock unloading container ships will haunt me.

Strange - same here. Not sure why Smart Crane is something I couldn’t under.

Had the same thing for a moment. Perhaps because I'm a not a native English speaker.

Modern container ports are already equipped by extensively smart cranes (although yes, they still require a skilled and highly paid operator). So it is unlikely someone would write such an article in 2021 :-)

A cane that doubles as a monopod for a camera would be cool. Something that can switch from a single tip to a tri tip tripod.

... for blind people???

Non blind people used to carry a cane around. I think more of a multitool in a cane format for regular people.

Right, but TFA wasn’t about those canes.

For the monopod case this is easy to find with hiking poles (a camera mount in the grip).

The tripod case is unlikely to be popular for reasons described in the article: it would add weight for a mostly uncommon use case, add something that could break (think about how you use hiking poles: I’ve bent quite a few, always in the lower segment) which would then make it deadweight you’d be dragging around for many days until you got out of the wilderness.

I've just realized that there are probably raging internet debates about what kind of cane tips are best. I'm not sure why I think I'm surprised by this realization.

As for "smart canes", I always figured it'd be more useful to have a fixed single-point LIDAR on a rigid head mount with some fast, imprecise output device like an eccentric buzzer motor with logarithmic amplitude for distance. Point head, get rough distance, just enough to get a general impression of how far away a wall is or where there's an opening. Apparently this isn't what people are talking about when they complain about worse-than-useless smart canes, though? It seems really obvious that speech synthesis might be the worst possible output system for something like this.

/ramble

"lidar for the blind" -> https://www.fastcompany.com/90574130/this-startup-wants-to-r...

Some people call it a https://twitter.com/hashtag/DisabilityDongle though.

Yeah I was thinking chest would be more ergonomic.

I can see why people don't like that one, yeah. It obviously can't replace the cane; you're always going to need at least one tool available that's as infallible as having a chunk of rigid stuff actually run into an obstacle or fall off an edge. It wouldn't even be useful in addition to a cane due to the awful ergonomics and interface.

What about fingertip mounted (rings, basically) small ultrasonic sensors? Each could be mostly independent (save for battery, I guess) and exploring the space would be as easy as slightly waving and tilting the hand. Tactile feedback would go straight to the fingertip. Quick to wear and remove.

Perhaps wrist bands would be more feasible with current battery/sensor tech, but, yeah.

The advantage of the fingertips is that they're sensitive and especially, they're extremely mobile. I imagine that if the sensors were mounted on the fingertips of a glove, on the back of the hand and pointed orthogonally to the plane of the hand, it would be fairly natural to develop a sort of "saccadic hand movement" to get a rapid perception of the surroundings.

Yeah, a glove would allow for much more battery capacity, of course.

Can't obstruct fingertips, those are necessary for the cane and braille. Glove might work, but I'd worry about it competing with cold-weather gear like mittens. Was why I was thinking of a rigid head-mount.

> Can't obstruct fingertips, those are necessary for the cane and braille.

Wear it on the left hand? Also, a glove is pretty quick to wear and remove.

> I'd worry about it competing with cold-weather gear like mittens

If it's useful, I don't think it would be an issue to create a "winter version".

> I was thinking of a rigid head-mount

If you hold your hand in front of your chest with the palm towards the chest, and wave it, you can appreciate how easy it is to scan the space in front of you. And the feedback would be directly on the fingertips, 1cm away from the sensor placed on the nail side of the finger, which makes for a very self-contained device. Also you get a free association between your self-perceived hand position and the space reading, which is nice. With a fixed head mount you need to constantly move your entire head to scan the space and the (tactile) feedback might be on an unrelated and less sensitive part of the body.

Reply on news.ycombinator.com