htns

I don't much follow AI news beyond what I randomly happen to see on HN, but this might still be the largest open source model: https://github.com/yandex/YaLM-100B . There's discussion of it here: https://old.reddit.com/r/MachineLearning/comments/vpn0r1/d_h... - at the bottom of that page is a comment from someone who actually ran it in the cloud.

The only time I tried to use Discord it demanded I send them my government ID.

The model of the game seems to be to have steep prices, but then give something like $100 of premium currency monthly in exchange for playing the game. This will generate engagement even if the content is mediocre. How many players would do dailies for fun if they did not come with what's effectively a $1 coupon?

I'm almost in the same boat, though I use emacs' language-specific input methods for longer texts. The postfix input methods especially work great with Dvorak.

They had a bit under $80k in crypto in their list of BTC and ETH addresses "leaked" along with the source code when the site was hacked earlier this year.

Probably one of the reasons Western Europe diverges from the US, despite the academic scene being similar and even more entrenched, is that the conservative bogeyman in Europe was smaller.

I don't recall ever having heard about Blokada before, but looking it up now it doesn't seem recommended: https://gitlab.com/fdroid/fdroiddata/-/merge_requests/8536

I'd guess anywhere snow can build up on roofs will have the lines buried.

This will be a nice case study in if collecting telemetry actually works - will audacity actually have improved in a year from now?

Why would anyone acquire it except to turn it into malware? I can't recall a project that wasn't built for commercial success from the start being bought for any other reason. I guess sourceforge stopping malware bundling could count.

Wayland will win because Xorg development was taken behind the barn years ago in anticipation of Wayland finally being ready.

There's Crystal: https://crystal-lang.org/

The CUA issue seems frivolous, since retention among those who don't even do the built-in tutorial on the splash page was never going to be high. It might help somewhat, but the bigger issue with emacs is the association with spacemacs (vi), lisp, and "configurability", which attracts some but perhaps leads many more to disregard emacs as Kool-Aid and never even try it. The reason to use emacs is that the text-editing is superior. magit, org-mode, email, and so on having an unified interface is great, but the text-editing (navigation, mark and region, macros, rectangles, and more) is what I miss acutely when using anything else, and it's also the reason why the rest of emacs exists.

Because governments are wiretapping the entire internet they can track down any server they can connect to live regardless of the protocol or number of indirections (this could be prevented with Freenet-style distributed hosting), but the actual downfall of the dark web seems to be web technology and outsourcing hosting. Last I checked Tor browser didn't disable javascript even on .onion sites (restricting HTML to a subset and requiring CSP would go a long way), and when sites share hosting they tend to go down all at once.

Tools like Dwarf Therapist might give the impression that labor assignments actually matter a lot in the game, but really workers are really productive and a full-grown fortress of 200 dwarves can be easily run by under 10 workers. You could assign all your labors in the embarking screen, assign the first few arrivals as farmers and miners, and all the rest can be left to idle, or drafted into the military, or made scholars or priests or entertainers.

The likely reason for there not being food is underproduction. Causes of that could be not having enabled planting for every season, planting textile crops or crops that need to be processed before they are edible (if the textile crop fields are too large it might be fine on embark when you have only 5 seeds, but end up using up all planting time once the seed count is high), lacking seeds, perhaps from cooking the plants directly (cooking does not give back seeds), or being bottlenecked on field size or planter count. Food will rot if not in a food stockpile, but the rate is pretty slow and easy to outproduce, also if underground the rot will produce miasma making it easy to notice. The next likely reason is that a single dwarf might have become unable to path to food. In such a case the dwarf is likely walled off from the fortress in either the caverns or the surface, because anywhere else there wouldn't be water and they would die of dehydration first. Disabling web gathering at game start helps prevent dwarves being stranded in the caverns. It's also possible all food was forbidden or there was an active civilian alert.

It's a bit complex in that there are many ways to fail, but again the winning formula is simple. For me the learning curve was first giving up because I didn't even know what I was looking at or that z-levels existed, and then a year or two later watching a video tutorial that quickly laid down all the basics, and finding the game relaxing to play after that. Unfortunately I don't remember what the tutorial was and some tutorials out there are pretty mediocre, e.g. anything that recommends making a refuse stockpile outside is not adequately conveying that sealing off your fortress from the outside world is key, and IMO building an indoors well should be mentioned as something you need to do even if the process's not covered.

Same. My go to has been to use Qubes if you at all can, because it's actually secure, and then to use Ubuntu, because it actually works. To me most of the bad reputation of desktop linux seemed to come from people refusing to use Ubuntu for demented reasons... But I have must not have been following distro news at all in recent years, because I only just now learned snap is not fully open. That's quite the cynical walled-garden power grab and bad enough by itself to drop Ubuntu.

Since late last year crystal made overflow checking the default (https://crystal-lang.org/2019/09/23/crystal-0.31.0-released....). Dunno how it changed the benchmark.

False of course implies true, but the deductions are supposed to be in reverse. So a chain of deductions could be P <= P' <= true where P is the original goal.

Crystal mostly copies ruby, so it's probably not that naive (though actually I don't know how ruby is on windows).

That's equivalent to wrapping an enum in a class. Emulations of type hierarchies without OO often fail like this, having a A-or-B be literally the same type so losing out on type safety/forcing constant rechecking of the discriminant.

A good UI is key. If your editor can navigate blame and commit history with just a key press, it actually speeds up figuring things out, especially in old codebases with bit rot. Benefiting from 10+ years of history might be rare, but two years feel like yesterday.

Exponential population growth leads into hyperbolic economic growth.

You do need to control both the key and the channel in this case.

I've never heard of an AGPL project being beset by entitled leechers... granted the license doesn't really matter with end-user applications.

There are a lot of people who never use a mouse while browsing the web! That's possible with browser extensions like Vimium, which show you a keyboard shortcut on every link when you press a key. "Two or more keystrokes" is a lot faster than using the mouse with that scheme.

A drawing tablet is better than a literal mouse. They are pretty cheap, have better ergonomics, and you can use them proficiently pretty much instantly.

I don't have any particular definitive solution in mind. Attestation is possible (e.g. https://developers.yubico.com/PGP/Attestation.html), but just telling people to do it right should go pretty far, especially if the packaging software doesn't cater to circumventing the policy.

Qubes OS already exists! It's been perfectly usable for more than a decade now. It can't be recommended enough: https://www.qubes-os.org/

Sadly "modern security awareness" doesn't seem to really amount to anything, and existing solutions go unused. %90 of this "malware library" problem too would have been avoided if package repositories just required all packages to be signed with keys on hardware dongles. Ruby and python and some others at least have the excuse of inertia, but it's pretty cynical of everything else to not require signing already, when the added friction is nothing relative to the effort of writing software worth publishing.

Think of it as memorizing 4 IPv4 addresses.

I don't have the exact command at hand, but it just uses tesseract. Something like "import - | tesseract <whatever options for language and page format> | xclip". The accuracy is not amazing by any means.