pointlessone

I wish people did more testing or at least acknowledged the limitations. I'd expected at list some level of diligence from Dave. This is half-broken in Firefox and looks like a regular nested list in Safari.

I understand it's made for personal use but if it's posted on the public web at least a disclaimer would've been nice.

There are a plenty of free LLMs out there. It doesn’t even have to be particularly smart. It can just infinitely nit pick and ask “why did you do it like this?” on random lines of PR.

I wonder why we haven’t seen AI weaponised in the opposite direction yet? Say, a bot with a prmpt like “criticise this PR into oblivion” let loose on AI-generated PRs. The “contributor” is supposed to address the concerns but will they be able to?

It’s not about trade. Trump and his administration pulled Ukraine aid on multiple occasions. They halted intelligence sharing for like a day and a half and that let Russia retake Kursk region.

Once a solid partner Trump turned USA into an unpredictable dependency that can change allegiance at a drop of a hat.

And "the best weapons” USA sells are best not because of hardware but because of the services that pull the hardware together. Patriots, F-35, even Abrams are all so good because they’re all networked and work together. If the service is cut it all become much more expensive and so much less usable than competition from Europe.

So while technically Mirages, Grippens, Typhoons, and whatnot are a bit less advanced than F-35, now they seem much more reliable.

Except the example of a system OP praises can not be ran locally. Well, it can but it’s a full on docker compose setup with half a dozen microservices.

I can’t say I like OP’s vision. My main objection is that this vision is terminally online. I want to be able to run the whole build locally (for when my internet is down, or I’m on a plane, or on a remote island in a cave, etc.). The local build and CI should only differ in that local build is triggered manually and results are reported in the terminal (or IDE) and CI build is triggered by a push and reported on the PR (or other web page, or API endpoint, etc. ). It should be the same but for the entry and exit. Tasks, queues, DAGs, etc. it’s all nice but ultimately are implementation details. Even make has DAGs, tasks, and parallel execution. Unless the build can run locally it’s as if there’s no build. Differences between local build and CI, be it because of environment, tasks setup, caching, whatever makes CI painful. It’s precisely because you have a build system for local builds and a separate CI setup that the world contains 10% more misery than it should.

So basically either the whole CI pipeline is just a single command invoking my build system or the CI pipeline can be ran locally. Any other arrangement is self-inflicted suffering.

These visualisations are so bland I can’t tell them apart. I mean, if looking at two side by side than sure, but if I look at them like every other day at best I can’t tell if anything has changed.

Here you go: https://indiandefencereview.com/scientists-organism-living-n...

I kinda do.

I keep a tab open if it’s something I want to look at later. I have like a bajillion opent tabs. Literal southands (plural). Once in a while I go through those and close a lot of them. Like hundreds at a time.

And I keep a web archive of a page in DEVONthink if already read it and I want to keep it for reference. Southands of pages spanning more than a decade back. Quite a few of them ar not on the web any more. Some are not on the Internet Archive either.

I feel for Codeberg and people who against AI but also I think Anubis can’t die soon enough. It breaks archiving and is very annoying when JS is disabled or when faced with an aggressive ad-blocker. It breaks web in more ways than one.

There’s dependency in the nginx config. You have to specify where your certs are. So you have to have a working config before you start nginx, then you need to get certs and change config with the cert/key location before you can HUP nginx. This is extremely brittle, especially if you have a new box or a setup where you regularly bring up clean nodes as that’s when you can get all sorts of unexpected things to happen. It’s much less brittle when you already have a cert and a working config and just renew the certificate but not all setups are like that. I can’t even confidently say that most are like that.

Yes, a few lines in the config and you don’t need certbot any more.

You can specify any ACME API base URL. It’s not just Let’s Encrypt.

DNS challenge is complicated by the fact that every registrar has their own API. HTTP is easier for nginx because it’s a single flow and it already does HTTP.

I’m sure nginx will get DNS but it’s still an open question when it will support your particular registrar or if at all.

Nginx with certbot is annoying to setup. Especially with HTTP challenge. Mostly because of a circular dependency. You need nginx to clear the challenge and once verboten gets a cert you need to reload nginx.

I switched to Lego because it has out of the box support for my domain registrar so I could use DNS instead of HTTP challenge. It’s also a single go binary which is much simpler to install than certbot.

You don’t need exactly the same cert for failover. You only need a valid certificate. You don’t even need the same cert for every entry in your load balancer. Client will pick a single IP address when resolved, then connect to it and will keep using that TLS connection for the whole session.

For all the shorcomings of AI and everyone involved this must be one of the silliest takes.

- The latest concept sports car costs a few lifetimes of income in low-income countries

- Aircraft carrier costs 200 GDP of Tuvalu

Now… any chances we could get a thing that takes in a React app and spits out a normal website with minimal or no JS?

100 km is like 12 hours. Probably much more if you’re just starting. 2h/day is quite a commitment.

The description is not too dissimilar from a badly photoshopped motel ad. Yes, it nominally resembles reality but just barely and doesn’t tell you about all the nasty things that come with the tiny pool.

Yes, everyone had a job. It was illegal not to have one. Jobs were basically assigned by the state. Most “normal” people could choose among a few OK-ish options. But if you got a hair out of line or crossed wrong people you could get assigned shoveling coal in a small boiler room or building railroads in Siberia. And if you didn’t want that job you became a criminal and went to some mine or quarry working double shifts for gruel in Gulag.

Yes, everyone had a roof over their head. But that doesn’t mean space, comfort, or even privacy. It wasn’t uncommon to have no more than a bed and a small cabinet or chest. Most of the living spaces and urban areas were extremely depressing. It’s all grey boxes and muddy trails. The best you could hope for in most places is some trees and bushes adding a bit of green in the summer.

There was very little people could do to change any of that. You couldn’t move elsewhere just because you wanted. People in cities with better professions could move if they managed to know the right people and could get a job elsewhere. People in villages were basically pinned to their colhozes. Their only option to see other places were a holiday trip if the managed to save up a little or to go into military and be stationed wherever.

This lack of agency resulted in mass depression and alcoholism.

Where do you think the grim and drunk russian stereotype comes from? It’s just the soviet reality.

I wonder how visually impaired to the degree that they need a tactile map would orient themselves on location.

PDF doesn’t have to be bad. Tagged PDF can represent document structure with a decent variety of elements, including alternative text for objects. Proper text encoding can give a good representation of all the ligatures and such. All of this is a part of the spec since 2001. The fact that modern software produces PDFs that are barely any better than a series of vector images is totally on the producers of that software.

Didn’t they promise GPT-5 this month? The one that got gold on IMO?

8 bits per channel. 8 * 3 = 24

Ah, I was wondering why would anyone tell they can get 5 metric tonnes of gold (~$535M) for 2.5 GW of power (~$500). Regular mercury is ~$210/kg ($1.05m/5 tonnes or 500x cheaper than gold). Although, Hg198 has 10% natural abundance. So maybe they can use raw mercury and still get decent returns depending on what othere isotopes decay into.

I moved from Pocket to DEVONthink a few years ago. It can save proper web archives and have quite a few sync options. It’s a little unconventional use for Dr but I love the experience.

MK CGI might as well be on purpose. Paying homage to the cheesefest of the old movies. The CGI in the Uncaged Fury trailer is definitley on purpose an intentionally worse than bad.

The dev, having access to source code, issues the final patch that disables that feature. Better yet, the dev, having access to source code, never implements this feature in the first place as it doesn’t improve the player experience in any shape or form.

Saas offerings are safe as they don’t let you Buy a thing. They’re very transparently a service. Same with timed licenses. It’s clear what you’re paying for. Most of the time.

With games it’s often not the case. You buy them and then at unspecified moment you might lose a significant portion of functionality or the whole game. At the time you buy a game you don’t know if it’s for a time, what time, or what will stop working.

If it’s manufactured it implies intent. Someone at Microsoft is doing it on purpose and, presumably, thinks it’ll benefit them. I’m not sure how this can be seen as a win for them. They invested a massive amount of money into buying all those game studios. They also admitted Xbox hardware is basically dead. So the only way they can any return on that investment is third party hardware: either PlayStation or PC. If I were to choose it would be pc for MS. They already have game pass and windows is the gaming OS. By giving business to Sony they would undermine those.

I don’t think nVidia wants gaming collapse either. They might not prioritize it now but they definitely know that it will persist in some form. They bet on AI (and crypto before it) because those are lucrative opportunities but there’s no guarantee they will last. So they squeeze as much as they can out of those while they can. They definitely want gaming as a backup. It might be not as profitable and more finicky as it’s a consumer market but it’s much more stable in the long run.

Wasn’t neurogenesis accepted since like 80s?