marklar423

Very well said, I often feel the same way.

About memories - throwing away the associated item often feels like admitting that you can't go back to the moment represented by the item. It's hard to face that small grief so it's easier to keep it there.

I agree but I think being usable by the keyboard isn't enough, because the shortcuts are often hard to discover and remember. I think the ideal is - like good TUIs - GUIs should put obvious hints on screen how to navigate via keyboard.

It would be really great to have some GUI frameworks for the common platforms (including web!) designed to do this and have some opinions on common shortcuts for common actions so we can standardize on something.

To me, the funniest part of this is:

> We had to change them all

I'm assuming a bot running a headless browser instance can still get past it?

It's still valuable to raise the cost of scraping of course. I don't think anything can really stop a determined scraper from impersonating a human. I wonder though if a system similar to Anubis but mining some crypto would make bots _welcome_ - since they're paying for their traffic.

This is a great article and exactly mirrors my experience consuming webhooks. My team built almost the same workarounds as the author.

The solution my team eventually settled on to the dedup, buffering, and race problems was:

  * When a webhook came in, store a single copy of the payload (usually json) in some temporary storage and enqueue the id.
  * Any subsequent updates (while the id was enqueued) overwrote that payload and skipped the queue.

So, we were able to ensure we only updated the object once and with the latest state (since the webhook payload always held the entire object state).

We didn't have a great solution for the bootstrap problem though.

I do like the SCROLL proposal, but I wonder about the cost of keeping around log-structured data forever - every log structured DB I know about does compactions for this reason.

> However, the sharp monthly increase needs context. Statcounter’s June figures included an “Unknown” category that accounted for 9.24% of North American desktop usage. Its reduction seems to have coincided with Linux’s rise, possibly indicating better identification of previously unclassified traffic.

Still good news for those wanting to see broader Linux adoption, but the news really is the usage was correctly recognized as Linux for the first time.

I stand corrected, thanks for pointing it out. I suppose then I'm back to not seeing the utility of the proposal over form tags.

I had the same thought at first, but the example in the article convinced me. It's something I ran into when I did webdev.

Form tags usually contain other elements - textboxes, checkboxes, etc. The url and method are specified on the form tag and not the button, so if you want a second submit button that goes to a different URL or use a different method, you need to duplicate all the other inputs in a separate form tag with your button.

All of this can be done with scripting, but the point of this proposal is to make that unnecessary.

Did Apple help defend you against those claims during the six months?

I've always understood it as a perk the company knowingly allows its employees who need to travel.

+1 thanks for porting this

Yes!!! I spent many hours playing the summer resort games.

Something about a pool party? I remember that one

> Alongside these games there have always been small groups of men moving around groups of small men in a basement somewhere re-enacting some battle or other.

Amazing line.

Just want to drop this link to the excellent https://acoup.blog/2026/02/13/collections-against-the-state-... which discusses the different costs of war, including how significantly weaker powers can win by increasing political costs.

Maybe I'm splitting hairs, but GP called it a copyright violation and my understanding is a "clean room" reverse engineering for interoperability was fair use and not a copyright violation.

Yes this does threaten Blizzard's business model so I understand why they'll go after Turtle, but that doesn't mean we have to care or let them prosecute Turtle for Contempt of Business Model.

Now, if Turtle used Blizzard's WoW trademarks to advertise and make money, I fully agree that violates their _trademarks_ and can be litigated as such. But if Turtle somehow didn't do that (and still sold access to their compatible WoW backend), I'd be interested to hear if that is somehow still a copyright violation.

If it turns out the private server code was a greenfield reverse engineered effort - do you still think that's a copyright violation? Why?

Cool! Thanks for the clarification.

In the discussion of the Triforce arcade compatibility, there's some discussion of "IC Card" support needing to be implemented, and doing so unlocking a lot of missing functionality.

I think this is referring to the Japanese rail payment cards? I know you can use them on things like vending machines, but from the article it seems like the Triforce cabinets let you save game progress on them too, which would be a great feature I've never seen in US arcades.

IIUC, if you have the source you can recompile said Windows app with LiteBox to statically link in the Windows OS kernel dependencies, so it'll run on any compatible processor regardless of OS (since it won't be making syscalls anymore). It's a unikernel basically.

That's the theory, but I don't know how far LiteBox is along to supporting that workflow.

Can tailscale connect to hosts behind CGNAT?

Would you mind pointing me at the research you found? I've been looking for studies that correlated hypoxia and autism (and related interventions that might help) but I haven't been successful.

Can you elaborate on why the raspberry pis are unreliable? Is it the SD cards, or something else?

There are certain genetic markers you can test for, but not all forms of autism appear in the tests we have today.

Then there's things like the folate blocking antibody (https://pmc.ncbi.nlm.nih.gov/articles/PMC4783401/) which you can do a blood test for, but again not all people with autism have the antibody.

Yes it's exactly the one-off situations like that, which aren't super often but occur enough to greatly inconvenience someone without a pocket browser.

I've been trying to do this too - paring down distracting apps, leaving only essentials like communication, maps, uber, etc. But my problem is what to do about the browser? I feel it's too essential to the "long tail" of uses (as the author put it), but also among the most distracting apps on my phone.

If anybody has any ideas I'd love to hear them.

Here's how I think about it:

- Money (the concept) is useful to society as a store of value, so you don't have to waste effort bartering for things.

- Adding on to that, credit is useful to society since it lets humanity even more efficiently allocate its good and labor (stored as money).

- Finally, stocks, insurance, and other financial instruments are additional advanced developments on top of credit, where groups of humans (companies) can take on even more risky endeavors supported by investors or insurers.

So my view is companies like Jane Street facilitate these complicated value transfers, to let (e.g.) a spaceship company draw on resources generated by growing crops, selling shoes, giving haircuts, etc via a convoluted path through stocks, ETFs, whatever.

I think Git was initially more popular and that had compounding effects that made it eclipse Mercurial.

In 2010 or so we tried to adopt Mercurial at the small company I worked at, but the support for Git was just so much better - even back then. Git's popularity meant that tooling, documentation, and general ease of finding people to ask questions was 100x easier with Git, so we switched. I'd imagine the same thing played out many times over.

A few FAANGs did and still use Mercurial, but they're on their own islands and don't really affect the wider tech ecosystem.

I'm self hosting Readeck (https://readeck.org/en/) and I really like it. It's nicer than Pocket was, the website extraction seems to work better, and it can't ever be shut down.

For my Kobo, I wrote a mod that lets me redirect Pocket API requests, and a small proxy server that translates Pocket API calls into Readeck calls.

So far it's working flawlessly and my Kobo is using its built in Pocket viewer for Readeck instead. I'm hoping to open source it soon so others can use it.

My company moved EiBs of data off of tape a few years ago. It was reliable and durable, but the problem was read speed.

It took so long to move tapes around and read the sequentially (no random access!), and as the data corpus grew it got harder to have a practical backup, even though the data was still theoretically extant.