HN user

sisk

605 karma

[ my public key: https://keybase.io/jeancharles; my proof: https://keybase.io/jeancharles/sigs/tHJQXHjedDoWcAmTSK-S-ZBPECxJ9CDBFqfh2V33Ba8 ]

Posts6
Comments134
View on HN

Re GPU usage: noticed this myself a few months back and decided to poke around. Recommend turning off the minimap if you have it enabled. There seems to be some over-rendering that is exacerbated by it. It's actively being investigated and improved upon (0.192 was a great step in the right direction) but, even still, I've found it (currently) has a big positive impact once disabled.

Hiya. No questions but just an fyi: the version of `@fresh/init` mentioned in the article (`2.0.0-alpha.30`) fails to run. Looks like latest at this time (`2.0.0-alpha.33`) runs successfully.

zsh also supports the "alias anywhere" concept (their term is "global alias") by using the `-g` flag.

   alias -g ag='2>&1 | grep'
   some-command ag 'words' # equivalent to: some-command 2>&1 | grep 'words'

Websockets are bidirectional while SSE is unidirectional (server to client). That said, there's nothing stopping you from facilitating client to server communication separately from SSE, you just don't have to build that channel with websockets.

For the first part of your question about the other replicas, docker will load balance between all of the replicas either with a VIP or by returning multiple IPs in the DNS request[0]. I didn't check if this proxy balances across multiple records returned in a DNS request but, at least in the case of VIP-based load balancing, should work like you would expect.

For the second part about updating the service, I'm a little less clear. I guess the expectation would be to bring up a differently-named service within the same network, and then `kamal-proxy deploy` it? So maybe the expectation is for service names to include a version number? Keeping the old version hot makes sense if you want to quickly be able to route back to it.

[0] https://docs.docker.com/reference/compose-file/deploy/#endpo...

The private key is much easier to set up than stunnel's TLS certificate, "dd if=/dev/urandom count=4 bs=1k of=key" and you're good to go.

The spiped documentation recommends a key size with a minimum of 256b of entropy. I'm curious why you've chosen such a large key size (4096b) here? Is there anything to suggest 256b is no longer sufficient for the general case?

Enabled by typing "max" in the Arc command bar (cmd + T).

Will be available for free for at least 90 days. Video did not mention what would happen after that time—they said they'll listen to feedback.

Features:

Ask on Page - Hold down Command + F on any page to ask a question and let Max answer it for you in seconds.

5-Second Previews - Press Shift and hover over any link to generate a summary of the webpage, without a single click.

Tidy Tab Titles - Have your tabs automatically renamed with tidier, shorter titles when you Pin them.

Tidy Downloads - Keep your many files more organized with smartly renamed downloads - and make them a little easier to find later.

ChatGPT in the Command Bar - Press cmd+alt+G, start typing, and get answers in fewer clicks.

(This summary generated by me, a real human person, NOT an LLM. Yet.)

Many of the gl.inet routers can run vanilla OpenWRT out of the box—the linked router above (Beryl) included[1]. Be mindful that not every one of their routers can as some run unsupported chipsets that require a custom build, but many do. Can always check for support on the OpenWRT page for gl.inet routers[2].

Just here to second the recommendation. I'm in no way affiliated, I've just happily used several generations of their routers for this exact purpose.

EDIT: Wanted to point out that their newest and most powerful travel router with (upcoming—in the latest v23 release candidate[3]) support from mainline OpenWRT is the Beryl AX (GL-MT3000)[4].

[1] https://openwrt.org/toh/gl.inet/gl-mt1300_v1

[2] https://openwrt.org/toh/hwdata/gl.inet/start

[3] https://firmware-selector.openwrt.org/?version=23.05.0-rc3&t...

[4] https://www.gl-inet.com/products/gl-mt3000/

To add, modern u-boot has enough UEFI support that you can get away with using it instead of this full tianocore-based implementation.

Loosely related: a few years ago I wanted a simpler alternative to some of the more feature-full log viewers out there so I threw together a tiny (50kb) app that might be useful to some folks in here.

All it does is consistently colors the first field in a line from stdin so you can quickly see which log lines have the same first field.

I used it in combination with the parallel[0] command to prefix log lines by replica name when tailing logs across machines: https://github.com/jasisk/color-prefix-pipe

[0] https://www.gnu.org/software/parallel/

In case you have multiple minimized windows, another seemingly little known trick is pressing 1 while in the application switcher which will show all of the windows for the highlighted application.

Sure, the _amount_ of content has grown given the lowered technological barrier for entry and the ease of distribution but content quality isn't some finite resource that is spread out within some limit. If anything, more content means more opportunities for truly great stories to come from corners of the world that might not have previously had that opportunity due to any number of reasons.

Plus, these channels have opened up more categories of film to wider audiences. Big budget, full length films and multi-season tv series aren't the only viable options anymore—some of the most talked about media in my circles over the last couple of years have been limited series. Try dragging a brilliant 8 hour limited series across an entire 23 episode season or trimmed down to a two hour film and it's a completely different story.

Brilliant film makers are still brilliant film makers and the number of studios willing to take a chance and fund them has never been greater. The traditional movie-watching experience is still here, it's just no longer the only option. I don't think declaring the movie business as dead is accurate, it has simply evolved and adapted. But from where I sit, this evolution has just given us more stories and more ways to hear them.

Just because you can distribute a story without a year-long PR blitz doesn't mean you can't tell a good story.

From the linked article, here is the full quote:

"The entire infotainment system is a HTML 5 super computer," Milton said. "That's the standard language for computer programmers around the world, so using it let's us build our own chips. And HTML 5 is very secure. Every component is linked on the data network, all speaking the same language. It's not a bunch of separate systems that somehow still manage to communicate."

Curious why the author declares Container Linux bullshit/unsuccessful? I thought it was a wonderful project that got better over its (brief) lifetime. The active/passive upgrade was an absolute blessing. If the declaration is it was unsuccessful because it no longer exists, Flatcar and Fedora CoreOS are pretty straightforward successors. fcos is basically functionally equivalent if you don't use rpm-ostree layers and if you use podman in place of any rkt containers. If the complaint is "systemd," that's fine—it's not the OS for you—but I don't think that makes it bullshit.

I didn't like having to move cl machines to fcos but it really wasn't that bad and I still get coordinated, active/passive upgrades. ¯\_(ツ)_/¯

Founder mentions on twitter pricing tbd though hopefully this year, considering $30/mo with a discounted annual plan. Discounts/scholarships for students, non-profits, and low income. Also mentions a $10k lifetime plan with a potential “code escrow” should they cease to operate (though acknowledges not expecting to get many of those—mentions expecting 10, ever).