How To Build A Smartwatch: Software | Hacker News Reader

How To Build A Smartwatch: Software

ericmigi.com

121 pointsteekert50 comments

For me, the best that Pebble has to offer is: "it is your machine, you can do whatever you want with it".

Apple severely restricts what you can install in the hardware you buy from them. Google will soon restrict the installation of Android apps not signed by Google. Microsoft restricts you to use your computer without a Microsoft account. John Deere restricts you from fixing your machines with parts sold by others. Espresso machine manufacturers restrict the capsules you can use in their machines. AWS makes everything incompatible and hard to migrate to other cloud providers.

They all follow the IBM business model: you buy IBM and end up fenced in blue in a walled garden that you can't escape.

I don't want that. I don't want to buy machines that come with a leach.

I like it for what it is, a well-built smartwatch with with a SDK, internet connection, and decent battery you can hack on over the weekend. It isn't going to replace your Apple Watch, but it is significantly more refined than some other offerings such as the PineTime, Bangle.js, etc.

(Nothing against those projects, I enjoyed them for what they were as well.)

Having owned and worn Pebble watches for the lifetime of the original company and then some, I think you would be _stunned_ at how many little things Pebble got right that others didn't. Personally, nothing has come close to these devices since the release of the Pebble Time.

PebbleOS was quite polished compared to the other RTOS out there. The animations were smooth most of the time, and responsiveness was great, the concept of the timeline is genius. The Pebble team made sure that it was accomplishing its main purpose (always displaying time, and time related events) really well, and then built everything around that.

I hope this revival will be sustainable and bring back some of that brilliance.

Got my duo a few weeks ago and my battery lasted about 3 weeks before needing a recharge. It does everything I need. Great little smartwatch

Yep, my duo arrived about two weeks ago at 76% charge and is currently at 18%; I've only ever plugged it in for a few seconds to verify that the charger works. Amazing battery life.

(The buttons, however, are atrocious - mushy, hard to press, and literally falling apart. I'll probably do the 3D-printed button mod but to advertise this watch as IPX8 water resistant is ludicrous. The first button press out of the box put a crack in the silicone.)

Did you get it in black like I did? My buttons also cracked practically right away (within a day), I suspect because the reinforcements were installed poorly (the buttons are VERY hard to press). It made the down button unusable.

But kudos to Eric and Claudio, they're shipping me a replacement (in white, which, as I understand it and as they said in their email, should be less susceptible to the issue, something about the white rubber versus black makes it less problematic). My only frustration was how quickly it failed, since I know it's a new-old-stock case.

Highly looking forward to the Time 2. I only stopped using my Pebble Time Steel when the battery life degraded to ~3 days (after about 6 years), used a Fossil Hybrid for a few years, now a Pixel watch. Measuring battery life in weeks will be a breath of fresh air :)

Pebble was exactly what I wanted 10 years ago and exactly what I don't want now. I'm very happy to miss notifications when my phone is in my pocket or bag and I don't care about fitness metrics anymore. However I'd love to leave my phone at home while still being reachable...

I do use pebble for (something like) that. I find that setting it up so they certain people can get to my wrist makes me check my phone less. So I get sucked in less.

Their libc is kind of a mess of various sources. I wonder what's going on there?

https://github.com/coredevices/PebbleOS/tree/main/src/libc

I loved Pebble back in the day, and Eric is a great guy and friend to entrepreneurs trying to build cool things.

I do wonder how a modern revival of Pebble will compete from a product perspective within the current landscape. Obviously there's the high-end Apple Watches, but there's also incredibly cheap and long battery life products from China that you can see on Aliexpress and similar. Fitness tracking is another related niche that seems oversaturated, unless you do something really unique in biometrics sensing.

So it seems like a hard market to get back into, curious where they take things.

I think they have a very clear niche: nerdy techies (like me).

The question is indeed if it's a big enough market to carry to the company. I hope so.

The key value of Pebble to me was its incredible C SDK that made it super easy to write custom apps for it. I remember way back I got full turn-by-turn navigation working on it.

They sold the whole production run of Pebble 2 Duos, and they are keeping the company small to be sustainable this time. I think they have a chance.

The Pebble software is second to none in nailing the basics. I'll definitely continue to choose Pebble over no-name brands on AliExpress.

> They sold the whole production run of Pebble 2 Duos,

They actually sold more than the whole run; I ordered one, and recently got an email informing me that they don't actually have the parts to fulfill the order.

I used a super-cheap Chinese smartwatch (Amazfit Bip S) for years and recently switched to the Pebble. The Bip's battery lasted forever and it did check a lot of feature boxes, but overall it was clunky to use and not in any way hackable.

I switched to a Pebble 2 Duo recently and while the features are comparable on paper (multi-week battery life, reflective display, basic health tracking, etc.), everything is just nicer on the Pebble. The software is thoughtful and fun and there are tons of third-party apps, so it can do all kinds of things the Bip could never do.

There really isn't a huge market for this kind of thing; most people, including nerds, want a watch with a brightly colored screen and tons of health metrics and service integrations. I imagine Pebble will stay a boutique brand this time around.

If there is market for long lasting watch, I think it is if it looks like a traditional round watch. Or if it can work as outdoors watch. Garmin is moving from transflective to AMOLED for better colors, and there might be spot for rugged, long-lasting, cheap watch.

I think Eric has more-or-less implied that they will probably make a Pebble Time Round successor (no doubt with worthwhile battery life this time, given how much more the Duo is)

One thing I don't understand:

> Pebble 2 Duo is sold out! We are not making more. If you want a Pebble, I recommend pre-ordering a Pebble Time 2 soon.

Is this supposed to be a collector's item? I'm not sure I'd want to invest in an ecosystem where damaging the device means I'm out or stuck waiting in line for replacement - with no guarantee the new device will be similar enough.

Pebble 2 Duo were reusing the existing stock of Pebble 2 housings and displays. This model was intended as a limited run from the beginning.

Pebble Time 2 are designed from scratch and expected to be still available after the pre-order batches have been shipped out.

Thanks for the clarification.

I just hope supporting this limited run model will not consume too much resources.

The Pebble brand name definitely helps them break back into the market. Even some of my non-techy friends recognize the name.

except for the apps for those cheap watches, all your data is uploaded to servers somewhere all the time I assume.

I heard that no one likes him because he has no morals, but that's just what i heard

Looking forward to getting my Pebble Time 2. Its been fun to play around building a watchface in the mean time:

https://developer.rebble.io/tutorials/watchface-tutorial/par...

thank you for this incredibly transparent and honest post.

The section on "Setting Expectations" (5 employees vs 180) is the most valuable insight. As an indie developer myself, I'm deeply curious: how does this new "sustainable" mindset (vs. the old VC-funded model) change your prioritization for the software roadmap?

Does it mean focusing 100% on the core functions and being more ruthless about saying 'no' to feature creep, which is something that plagues so many other wearable companies?

Rooting for you all.

Does anyone know whether more than one device can connect to a Pebble watch at the same time? I'm thinking using it with your phone but also sending notifications from your laptop.

Pebble watches can use Bluetooth LE and Bluetooth Classic connection profiles simultaneously. It's possible to pair two phones this way, but that's considered an undocumented hack. Also there's no ready-made desktop OS support, I'd look into forwarding your laptop notifications to a phone via KDE Connect or something like that instead.

Connecting to a laptop isn't supported AFAIK. Personally I don't have any important notifications on my computer that don't come to my phone already. What would you use it for?

I'm thinking of having Claude send me a notification when it needs my input. I already do that sometimes with desktop notifications and this way I could water the plants or hang the laundry while waiting.

That's a cool idea! I think there are lots of ways to forward that notification to your phone. Probably vibe-codeable.

Can't be done, yet.

But I do want to dig into that, someday. There is an open source library in Kotlin multi-platform for building applications that interact with the watch (libpebble3) and, in theory, Bluetooth LE can connect to more than one device. But the PebbleOS probably restricts it.

My dream is to use the watch to authenticate to computers, websites and IoT devices.

PebbleOS is open source now, so there are no roadblocks preventing anything that is technically feasible.

I will buy it if it supports esim so I can be reachable by phone and sms all the time, which means some cell phone operator needs to buy in first.

I don't think being a phone-substitute in that way has ever been a goal for Pebble. It doesn't even have WiFi.

Its goals are battery life and a simple set of features including notifications from the phone via Bluetooth. You seem to have different needs, and there are other smartwatches being produced which attempt to address them.

Or "Milking content out of the old dry Pebble well"

They shipped hardware! They successfully open sourced a beloved platform and embraced the community ecosystem that sprung up around it. What more do you want?

They do business in a very sleazy way

Can you elaborate? I am hesitant to support a sleazy company, and am not aware of the behavior you suggest.

Your reply is an example of sleazy

Yeah, I got burned once, I don't want to be burned again. You will sell out at first opportune moment. Which is why I am passing on this. (I don't blame you for that, just don't want to be part of it)

BTW, Amazfit, rules.

I was bummed to see recently that Amazfit has stopped making watches with reflective displays! They're all OLED now just like everyone else's.

This is a rather bizare take. Pebble turned down a massive deal to keep doing their own thing. They sold cheap after they were going down because they had too mich staff, and not enough sales. Which Eric has said many times, and can even be found on his blog.

Then, when they were being sold, instead of shutting down the Pebble store and basically bricking all Pebble watches, they intentionally opened it up to make it possible for community support. Which is where Rebble stepped in.

Bizarre and disingenuous take. That really doesn't take into account Pebble's actions, much less their words.

What's bizarre? Per your own statement, "They sold cheap" as soon as they encountered some hardship, so it is quite understandable to not trust they'll behave differently this time around.

> "They sold cheap" as soon as they encountered some hardship

Nobody is perfect, and running a small hardware startup is difficult. I'm not saying Eric and co are perfect, but it seems like he's been fairly forthright about the mistakes made at Pebble[1] and what Core aims to do better.

Shit happens, people make mistakes, Apple/Google decide to compete with you and/or lock you out of parts of their garden.

[1] https://ericmigi.com/blog/success-and-failure-at-pebble

As far as I recall, it was sell cheap, or collapse with nothing. Garmin bought cheap and gutted it for the IP. It wasn't a sell to get paid, it was a sell or get nothing. It wasn't just hardship, it was the end.

It wasn't Garmin. Fitbit bought Pebble, then Fitbit failed and was bought by Google. Google then open sourced the Pebble source code.

https://9to5google.com/2025/01/27/pebble-smartwatch-2025-goo...

It was Fitbit, but otherwise, yeah

Amazfit is closed, Pebble is open. That's reason enough for me to avoid the former.

As enshittification encroaches on every corner of the technology ecosystem, a company putting out products in 2025 in a way that embraces its community and works in the open is laudable.

Maybe Pebble will turn evil one day, but at least the watches we have today will still work until they physically wear out, not when the company decides they should die.

How exactly did you get "burned" ? I'm still wearing an original pebble. It's worked fine for years.

Reply on news.ycombinator.com