Hacker Newsnew | past | comments | ask | show | jobs | submit | 77ko's commentslogin

Its on OR - but currently not available on their anthropic endpoint. OR if you read this, pls enable it there! I am using kimi-2.6 with Claude Code, works well, but Deepseek V4 gives an error:

`https://openrouter.ai/api/messages with model=deepseek/deepseek-v4-pro, OR returns an error because their Anthropic-compat translator doesn't cover V4 yet. The Claude CLI dutifully surfaces that error as "model...does not exist"


The config is really well setup - i am using Omarchy on a second pc (main one being a mac). Some thoughts:

DHH has good taste - leaving besides application choices (some of which I changed, e.g <insert_browser> instead of Chromium, no 1password), the configuration defaults all make sense (coming from a mac) - especially the key bindings.

Arch Linux by itself is a bit scary and requires config to make it "nice" so basically Omarchy takes away all the choices and config learning / pain - this tweet is a good summary:

> I've poured in endless hours configuring Hyprland + Arch, GTK/QT theming/scaling, auxiliary apps, and more to give you a superb base that can either be taken as-is or used to keep tweaking.[1]

Tiling window managers are great - I have young kids using computers for their hw and they prefered this over mac - windows. Which suprised me as personally it is a much bigger change for me after decades of regular windows/mac window management.

Linux / Hyperland Pros:

- I had a old pc from 2014 - which I put a minimal fresh new install of windows 10 - and it has been dog slow enough that it was waiting to be replaced. After installing Omarchy (Arch + Hyperland) it's perfectly fast and usable.

Cons: 1. Its designed to be a single user setup - the idea being u use HD encryption and login straight to the one true user. So for a shared pc its not ideal - the way its currently configured I think you need to run the omarchy bash install script for each user and also update individually for each one - not ideal for a pc shared with kids.

Really interested to see where Omarchy ends up. Its also given my usability ideas for my mac.

[1]: https://x.com/dhh/status/1932130355663761794


Sounds very cool! Are you displaying this on a dedicated screen or using an app on a lg tv?


Just using the TV web browser. Do you have kids? Mine are so easily distracted in the mornings. It spawned from the bewilderment of them having so few jobs to do in the morning (compared to us parents) and the complete lack of focus


uv is excellent! The only think I'm missing is an easy way to update all packages in an env, something like `uv update --all` or `uv update plotly`.

Which would fit in with existing uv commands[1] like `uv add plotly`.

There is an exisiting `uv lock --upgrade-package requests` but this feels a bit verbose.

[1]: https://docs.astral.sh/uv/guides/projects/#creating-a-new-pr...


Are you looking for something like `uv sync --upgrade`? This one should be re-assessing your dependencies (excluding version pinned ones of course) and regenerate the lockfile if I remember correctly.

https://docs.astral.sh/uv/reference/cli/#uv-sync--upgrade


Garmin Bounce[0] for kids, which is the watch which directly competes with this, has a 12-24hr battery life on LTE, although Garmin claims 2day battery.

[0]: https://www.dcrainmaker.com/2023/01/garmin-bounce-activity-t...


You are absolutely right. I'm certainly not pro-Garmin here. Just anti-12 hour battery life for kids watches. It's just not compatible with the way young kids work.


Reading the comments to dcrainmaker's review, his kids' Bounces were getting 36 hours of charge whereas several commenters state they were barely getting 10 hours, certainly less than 24 hours of charge.

Seems like the energy used to keep the LTE connection up is the main problem, if all the users have the similar settings to minimize battery used by the Garmin Bounce.

If Google/Fitbit says the ACE LTE only lasts 16 hours, I'd hope their estimate is much more conservative than Garmin's and has a lot less variance. Otherwise, I can't see the ACE LTE lasting until end of school day.


I only have to charge my kid's bounce every 2 days, 2.5 days usually. I do not turn on the "Live Tracking" very often but location updates, boundary fence notifications, voice/canned messages work great.


My kids main demand is the following features, and this watch doesn't deliver on any of them:

1. Whitelist contacts for calling and text call - installing an app is a pain for older ppl / friends.

2. Music: Why not Google? This already has headphone support and the Pixel watch hardware which support musics.

3. Maps: in case they need to go A to B. WearOS has google maps, so this should be an easy add. School bus is on google maps for e.g, so being able to check time to leave would be great.

4. Battery: 16hrs at launch is not going to age well...

Overall, despite being in the market for this, would not buy.

Features kids didn't ask for and I don't want:

Gaming: There is a market for gamification, but it seems to me the product team went overboard and spent way too much time here at the cost of making a better kids watch. I have no doubt it can and will make some kids more active, but...


Kids have never needed a watch to be active, run around and imagine creative games. Although, sure, we did spend hours during breaks all surrounding the one kid with a gameboy.


> 4. Battery: 16hrs at launch is not going to age well...

Sounds like a nightmare having to also manage your kids' watches and keep them charged just for them to barely make it through the day.


If they don't care enough to keep them charged, maybe they don't want/need them?


Good points! I can't agree more with item 1.


This looks amazing! I really like the clear seperation of loading/prepping data, and presenting it.

Some requests:

Add simple examples and more clarity to the publish docs.

I assume most ppl would prefer to deploy via github actions[1], which the docs currently just link to a complex deploy file - can you please add some more documentation on this, or add an example of the simplest possible deploy file?

Suggestion: is it possible to use a interface (like in vercel) to connect to a github repo and build/publish on changes?

[1]: https://observablehq.com/framework/getting-started#deploying...


This was a good interview with her: https://josephnoelwalker.com/147-katalin-kariko/

She had a very interesting life, I hope she writes a memoir.

Edit: She has a memoir out 10th Oct: https://www.penguinrandomhouse.com/books/706251/breaking-thr...


In this very moment, Penguin is getting quotes on getting "Nobel Prize Winning Author" stickers for the books.


I have a special hatred for "Now a Major Motion Picture" littering book covers.


Even worse is when they put a picture from the movie on the book cover


her journey to mRNA was full of adversity too. the NYT podcast with her was great coverage of this: https://www.nytimes.com/2021/06/10/podcasts/the-daily/mrna-v... (my excerpt https://x.com/swyx/status/1490363488824627200?s=20)

i cant imagine how it must feel to have her own vaccine injected in her, and know she saved so many lives through her persistence and the rare few that believed in her through her lowest moments.


Fantastic interview. Always love great branching off points from HN.


I've been using https://nextdns.io/ - works fast and most importantly blocks a bunch of adds (user configurable), so makes browsing on mobile much nicer.

Ancedotally the interenet seems faster.


Love it! Discovered something interesting very quickly. Bookmarked for future use.


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

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

Search: