> Initially, we experimented by building this as an app on Pebble, since it has a mic and I’m always wearing one. But, I realized quickly that this was suboptimal - it required me to use my other hand to press the button to start recording (lift-to-wake gestures and wake-words are too unreliable). This was tough to use while bicycling or carrying stuff.
I’ve never had that issue with my Apple Watch. Granted, apple are a world-class developer (arguably), so their stuff might be more reliable, but I use raise-to-speak and hey siri with my watch all the time.
This is probably also part of the reason that the AW battery lasts so much less time. It's always on the lookout for these events, and over time that ends up having a non-trivial impact on battery. For Pebble users, that would significantly compromise one of the main selling points of the Pebble.
But you have to instruct Siri to record/transcribe/save what you want to say; this does it automatically. It removes friction if you just want to record short notes.
I almost never want to take a note though, I usually want to perform an action.
Recording a note isn’t high friction in my opinion though: “Hey siri make a note XYZ”. Admittedly I don’t create or use notes like this, but I use reminders a lot and I’ve never felt like there was friction: “hey siri remind me to call Dave when I get home”
I do this too, the biggest issue I have is when the shitty voice to text doesn’t get it right, and I look at my shopping list the next day wondering wtf an “ear pig” is.
I feel there may still be a way solve it using existing hardware -- rather than go build entirely new hardware for it.
Some low effort ideas I cna think of -- a wake word instead of button press; a flick of the wrist; or maybe press the watch to your chin; There must a few more elegant solutions possible if smarter people put their minds to it.
Pixel watch is an order of magnitude more powerful and expensive than a Pebble though. Raise to wake is not as simple as it may seem considering most big brand smart watches didn't have a decently working implementation of it until only recently. It seems like the author wanted to keep it within the Pebble realm.
The pixel watch can be a nightmare. I've had the version 1 and the version 2 for years and I gave up because of the bugs. Random crashing, randomly dialing 911, just weird stuff that should not happen.
And then there's the support which is zero support. Completely frustrating to post a message and then get some volunteer support Tech, hahaha, saying expect improvements! And it's a volunteer saying that. And they have no authority. And There is no support, it all falls through. Random crashing on both versions of the watch. The first version screen was flickering like an old school television trying to tune in a distant UHF broadcast. Display drivers anyone?
So, pixel watches are in the drawer and I've got a Garmin watch on right now. Garmin is clunky but at least it's reliably clunky.
So it feels like a mis comparison, to me who's had the pixel watches.
I used to own the pebble, a couple versions of it, when they were first announced for several years again. And I found them to be very reliable and lightweight and usable.
I wanted a smartwatch that could talk to Google's home ecosystem and so I traded out of Pebble. And it's just been kind of mediocre misery.
Plus I don't know what Google is doing but recharging the pixel watch every 18 hours, or 36 if you're super lucky and your apps on the pixel watch behave themselves correctly, makes me feel like a slave to Google's naive product manager aspirations.
Like, "it can do everything, and we make money off of you because you are the product!" While at the same time making me miserable.
Economics don't work like that though. Google has the benefits of economies of scale and stable manufacturing partnerships, not to mention probably a lot more of the watch parts are designed in-house. I'd be surprised if Pebble could achieve anything close to the Pixel hardware for even twice the price.
It seems like Pebble is focusing on a niche market and this new product seems completely in line with that. There are plenty of other companies targeting the common folk who have no desire to optimize their life like this.
Niche market seems like an exaggeration, because they're competing against literal monopolies.
Pebble serves those people who want to watch or a ring that doesn't require being a slave to a wall wart, who want the watch to last for a long time. Take a look at Garmin, they do that too and they are a successful company. They are much older than Google and they still have a hard time keeping up with Google and it's billions of dollars of of mystery money advertising revenue.
The pixel hardware is a battery draining nightmare, in my personal experience having pixel watches for years and being a long-term pixel phone user. Even today the pixel phone that I have, after having I think five of these things, drains battery probably 20 to 40% faster than .. the competitor that I would next buy if I weren't feeling like I wanted some of the features that Google has bundled in with their phone and home and other like mnvo and messaging products. So, it seems like a mis comparison there, in my opinion. I don't want a smartwatch that lasts 25 hours and then has to be recharged. Or a smartwatch where the screen turns into a UHF channel just going out of tune and there's no tech support on Earth literally that is willing to help me. Volunteers on Google's support forums are lying to themselves that they have any power or sway with Google. It's a waste of time in my experience.
Man, I wish this was true. I've given the same feedback on a colleague's clearly LLM-generated PRs. Initially I put effort into explaining why I was flagging the issues, now I just tag them with a sadface and my colleague replies "oh, cursor forgot." Clearly he isn't reading the PRs before they make it to me; so long as it's past lint and our test suite he just sends the PR.
I'd worry less if the LLMs weren't prone to modifying the preconditions of the test whenever they fail such that the tests get neutered, rather than correctly resolving the logic issues.
We need to develop new etiquette around submitting AI-generated code for review. Using AI for code generation is one thing, but asking other people review something that you neither wrote nor read is inconsiderate of their time.
I'm getting AI generated product requirements that they haven't read themselves. It is so frustrating. Random requirements like "this service must have a response time of 5s or less" - "A retry mechanism must be present". We have a specific SLA already for response time and the designs don't have a retry mechanism built.
The bad product managers have become 10x worse because they just generate AI garbage to spray at the engineering team. We are now writing AI review process for our user stories to counter the AI generation of the product team. I'd much rather spend my time building things than having AI wars between teams.
Oof. My general principle is "sending AI-authored prose to another human without at least editing it is rude". Getting an AI-generated message from someone at all feels rude to me, kind of like an extreme version of "dictated but not read" being in a letter in the old days.
At least they're running the test suite? I'm working with guys who don't even do that! I've also heard "I've fixed the tests" only to discover, yes, the tests pass now, but the behavior is no longer correct...
I learned recently our company has an informal policy to discourage hiring fresh grads, which would explain why I haven't seen any hires at the bottom of our grading scale. I'm one step above (though arguably I could be two up by experience, I misplayed my negotiations), but I kinda feel like I'm dangling off the edge.
Until it isn't because enough people realize it is inherently flawed and will never result in stability and they would rather die than relegate themselves, their children, and all their later generations, to indefinite servitude of an outdated economic system. Provided we don't kill our world first in our pursuit of endless profit growth.
> I'd much rather deal with some less stability and have a lot more resources
This strategy works so long as you're in the group of people who do indeed have "more resources," but breaks down if you're marginalized in any way, or if you don't want to support the exploitative system.
That is what people always says about their current culture or economy or political system, even as they are on the very edge of systemic change. And quite frankly to me that just reads like an excuse to not do or try anything so people can kick the can down the road, benefiting themselves at the expense of many others both currently and in the future as things deteriorate further.
it is if that's all people do, but I think it is a necessary step to realize that we do live under a system that encourages an amoral allocation of resources. If people only ever shrug that off or treat it as intractable it will never change. Some would argue that it cannot be changed, but I think that just means they haven't read history.
I was talking to a friend of mine about a related topic when he quipped that he realized he started disliking therapy when he realized they effectively were just teaching him coping strategies for an economic system that is inherently amoral.
> So practically speaking, the options are follow the market or find a different line of work if you don’t like the way the market is going.
You're correct in this, but I think it's worth making the explicit statement that that's also true because we live in a system of amoral resource allocation.
Yes, this is a forum centered on startups, so there's a certain economic bias at play, but on the subject of morality I think there's a fair case to be made that it's reasonable to want to oppose an inherently unjust system and to be frustrated that doing so makes survival difficult.
We shouldn't have to choose between principles and food on the table.
> We shouldn't have to choose between principles and food on the table.
I am increasingly convinced that these are the only true kind of ethical decision. Painless/straightforward ethical decisions that you make every day - they probably don't even register on your radar. But a tough tradeoff does.
I don't buy this. Watching hi-res remasters of old films (Alien, Catch Me If You Can, Jurassic Park) you get incredible definition without damage to suspension of disbelief.
I think this comment[1] above really highlights what bothers me about modern shows and movies, but in short it's not the high def, it's how it's used.
I've ranted about this so much to friends, it's so distracting to me.
I have the hypothesis that some of it is due to how easy (or easier?) it is to quickly review footage to tell if the focus is set right when it's so aggressively shallow, but regardless of the "why" the monotony of it drives me crazy.
I want to see the environments the characters are in, I want the visceral grounding of the feel of that environment. I watched "Portrait of a Lady on Fire" and it felt so primal, and watching how intentional the sound design alone is I can see why.
> In such environments (visual and aural), nothing seems real and nothing seems like it matters.
That's a great description about the strange disembodiment I've felt in modern shows for a while now.
> Lots of tips on how to do this out there but one thing I do is have it try, throw away everything it it did, and try again with a completely restated question
This is the thing that worries me about AI/LLMs and how people "profess they're actually really useful when you use them right": the cliff to figuring out if they're useful is vertical.
"You’ve still come out ahead even if you take over fully."
I just finished a weeklong saga of redoing a bunch of Claude's work because instead of figuring out how to properly codegen some files it just manually updated them and ignored a bunch of failing tests.
With another human I can ask, "Hey, wtf were you thinking when you did [x]?" and peer into their mind-state. With Claude, it's time to stir the linear algebra again. How can I tell when I'm near a local or global maxima when all the prevailing advice is "I dunno man, just `git reset --hard origin/master` and start again but like, with different words I guess."
We have studies that show people feel like they're more productive using AI while they're actually getting less done [1], and "throw away everything it did and try again" based on :sparkle: vibes :sparkle: is the state of the art on how to "actually" use this stuff, I just feel more and more skeptical.
It was intended as a sort of military SF anothology, with the word "military" also encompassing revolution, insurgency, and various other 4GW concepts. The stories spanned the period from roughly 1980-2020, with many published from 2010-2020. The list was:
> A Dry, Quiet War by Tony Daniel
> ZeroS by Peter Watts
> A Soldier of the City by David Moles
> The Beast Adjoins by Ted Kosmatka
> Lady Be Good by Jack Campbell
> Mid-Death by Alan Dean Foster
> Weaponized Math by Jonathan Brazee
> Prayers on the Wind by Walter Jon Williams
> Highwaymen by Ken MacLeod
> Second Skin by Paul McAuley
> The Dread And Fear of Kings by Richard Paul Russo
> Herbig-Haro by Harry Turtledove
> The Lost Dorsai by Gordon Dickson
> Cincinnatus by Joel Rosenberg
> The Proud Foot of the Conqueror by Reginald Bretnor
It's the stories lower on the list that were unobtainable. Dickson, Rosenberg, and Bretnor being dead (may they all rest in peace,) did not help matters.
reply