QI said he and his team used several conditioning methods, but none of them were actually ringing a bell.

Plaid effectively admitted to stealing your transaction history and selling it to the highest bidder in the past. There was a settlement and they agreed to not to that in the future

Just don’t ever share your password, and certainly not your banking password, and definitely not with Plaid.

I also generally prefer a Condorcet Method (ranked choice, single winner) over mixed-member-proportional, but either one would be a massive improvement over our current system.

I’ll take Approval voting, even.

Too redundant, just use S-exprs.

(Mostly joking, but in some cases…)

Most people never become auto-didacts. Most auto-didacts still benefit from formal training because above average gross performance can mask subtle mistakes until the mistake becomes root cause for a significant error.

Under significant pressure (like a well-written dramatic fiction, but almost never IRL), most doctors will be willing to perform a procedure without formal training, but under normal conditions, they know it is not worth the additional risk.

Some people hate it, including some independent developers. I wouldn’t mind going without it, if there was a Free Software library management alternative. I want something to track what I have installed (because I’ve “lost” things and reinstalled them before) and something that has a decent uninstall.

I also get some benefit from the store integration, but I can understand developers being annoyed at the 30% “steam tax”. I’d gladly purchase using some other method, if I didn’t have to sacrifice library functions from previous paragraph.

It’s an argument against a license that permits relicensing under a more restrictive license. (E.g. BSD)

I read a story of someone that contributed to a BSD project, including fixes over some period of time, but later they ended up having to use a proprietary UNIX for work, that included their code, in a an intermediate, buggy state, but they were legally forbidden from applying their own bug fixes!

At the very least the GPL guarantees that if I am ever downstream of myself, I has fix my own damn mistakes and don’t have to suffer them.

I am still willing to contribute to BSD stuff, but vastly prefer something like the AGPLv3.

Hmm, maybe next time I’m buying games, I’ll pick up BL3 hoping it works on my Debian system through Proton or something.

Thanks for the info.

From the reviews I’ve seen (Outside Xbox), that’s the most you’ll get out of it. It’s mostly a nothing burger, “bad” is the best thing you can say about it.

BL2 still works on Linux, too. I can’t play BL3, which I hear is the better game (albeit with worse plot).

It’s been a LONG time since I tried, but ISTR it complaining that it couldn’t move the old_root because it was in use? But, when I’d fuser it, the only think that was using it was my shell (using it as the current directory).

I’ve never gotten pivot_root to work well / as documented. Major props for being able to pull this off. Hopefully you didn’t have much time where a connection-drop would require a reboot. SSH in the new root on an alternate port before the pivot? idk.

Given that some films are getting post-release edits before they appear on streaming or physical media, it might be a good idea just from an archival perspective.

I’m not a good judge of piracy methods for consumption, as I am currently willing to pay their prices or do without.

Depends on the pay differential and other options. I think it’s less useful for positions in my career, but it’s not an absolute no.

Might check out the Haskell layout rules.

Basically, when you leave out the ‘{’ then Haskell uses your intendation to insert ‘;}’ on later lines between the leading whitespace and the first token.

There some really old Haskell code out there that lines up the ‘{;}’ characters on the left under block-introduction keywords.

Actually, unless we want to adopt and propagate the Eq constraint, we can’t normalize in embed. Maybe it would be worth it to have a normal proof.

Which functions would you want to see?

– https://byorgey.github.io/blog/posts/2024/07/18/River.html

Since it’s a (uniformly) recursive data type, I want to see the underlying functor

data RiverF x r = CF !x | ConsF !x !r deriving Functor

projection and embedding functions:

project :: River a -> RiverF a (River a)
project = \case
  C x -> CF x
  Cons x r -> ConsF x r

embed :: RiverF a (River a) -> River a
embed = \case
  CF x -> C x
  ConsF x r -> x ::= r -- Choosing to normalize, here.

And then generalized fold / unfold:

fold :: (RiverF a r -> r) -> River a -> r
fold alg = f
  where f = alg . fmap f . project

unfold :: (s -> RiverF a s) -> s -> River a
unfold coalg = u
  where u = embed . fmap u . coalg

Note that since we chose a normalizing embed, the result of an unfold is also always normalized.

I would also not be opposed to a alternate definition that took a proof of normality for the Cons, but I know that would take some higher-rank types, at least.

ghciwatch (https://mercury.com/blog/announcing-ghciwatch) sounds cool. I’ve been using HLS and ghcid for work.

I have been in meetings which people who thought the fact that a user could use a different font, even only intentionally, was “unacceptable”.

I hope those people aren’t directing the ship at YT, but could be.