I can think if plenty of situations where system time is

• Optional
• Unreliable
• And even potentially disallowed by the user

In fact, if you don’t set up your containers right, the system time is almost always wrong.

If that’s the only error mechanism, sure. Exceptions in most languages tend to be relatively expensive, though, and most have a cheaper idiomatic way of returning error codes; you’d want to use those if they’re available, right?

Does Rust use exceptions a lot? I don’t know. V has panic and catch, but you almost never see them. Idiomatic is Option (?) and Return (!) values, which I thought V borrowed from Rust. Go does the (val, error) tuple-ish return thing, and while it too has catchable panics, they’re discouraged in favor of (error) return values.

Depends on the language. “Higher level” is a pretty broad field!

At first it wasn’t an issue: I used Voyager for this account, and Interstellar for the alt. Then I decided I liked Interstellar’s interface more and started using it for both. Both list the account in most places, but Interstellar doesn’t show it when replying.

I started making enough mistakes that I played with the settings and discovered Interstellar links the color theme to the account, and now I can easily tell which I’m using.

I’m certain I’ll continue to make mistakes. Thorn is surprisingly seductive, but the real issue is that auto complete and autocorrect on my phone keyboard has decided that the correct spelling for “the” is “þe”. I could correct it, but I feel bad for it; it’s just trying to he helpful.

Gosh darn it, am I using thorns in this account again?? I didn’t mean to.

I recently learned that only Icelandic does that. Eth was dropped early in old English, and thorn was used in both places. Additionally (as I understand it, now), while thorn was a direct “th” (voiced or unvoiced) sound, even when eth was in use it want orthographically a simple replacement for voiced “th”.

I guess Icelandic kept it, but eth was not in use through most of the old English, medieval period. And then the Normans came, and fucked written English completely up.

You’ll be absolutely thrilled to hear that I discovered that I can assign different color themes to different accounts in my mobile app, so these sorts of crossover mistakes should be greatly reduced.

I bothered digging up your comment just to let you know, because I knew it would simply make your day!

Toodles!

My recommendation is to put all of the variables in an environment file, and use systemd’s EnvironmentFile (in [Service] to point to it.

One of my backup service files (I back up to disks and cloud) looks like this:

[Unit]
Description=Backup to MyUsbDrive
Requires=media-MyUsbDrive.mount
After=media-MyUsbDrive.mount

[Service]
EnvironmentFile=/etc/backup/environment
Type=simple
ExecStart=/usr/bin/restic backup --tag=prefailure-2 --files-from ${FILES} --exclude-file ${EXCLUDES} --one-file-system

[Install]
WantedBy=multi-user.timer

FILES is a file containing files and directories to be backed up, and is defined in the environment file; so is EXCLUDES, but you could simply point restic at the directory you want to back up instead.

My environment file looks essentially like

RESTIC_REPOSITORY=/mnt/MyUsbDrive/backup
RESTIC_PASSWORD=blahblahblah
KEEP_DAILY=7
KEEP_MONTHLY=3
KEEP_YEARLY=2
EXCLUDES=/etc/backup/excludes
FILES=/etc/backup/files

If you’re having trouble, start by looking at how you’re passing in the password, and whether it’s quoted properly. It’s been a couple of years since I had this issue, but at one point I know I had spaces in a passphrase and had quoted the variable, and the quotes were getting passed in verbatim.

My VPS backups are more complex and get their passwords from a keystore, but for my desktop I keep it simple.

I started using the same client for both by “normal” account (this one) and my toy account (my pþþþt one) but have discovered that now it’s impossible hard to tell which one I’m in once I start replying. And I flip between them often, so now I’m accidentally posting eths and thorns here, and forgetting them more in the other account.

It’s a conundrum. I’m losing sleep over it, really.

I hope this isn’t a step towards replacing the native app with an SPA.

That’s the broken behavior I see. It’s the evidence of a missing understanding that’s going to need another evolutionary bump to get over.

Yeah, for me it’s more that just “produces correct output.” I don’t expect to see 5 pages of sequential if-statements (which, ironically, is pretty close to LLM’s internal designs), but also no unnessesary nested loops. “Correct” means producing the right results, but also not having O(n²) (or worse) when it’s avoidable.

The thing that puts me off most, though, is how it usually expands code for clarified requirements in the worst possible way. Like, you start with simple specs and make consecutive clarifications, and the code gets worse. And if you ask it to refactor it to be cleaner, it’ll often refactor the Code to look better, but it’ll no longer produce the correct output.

Several times I’ve asked it for code in a language where I don’t know the libraries well, and it’ll give me code using functions that don’t exist. And when I point out they don’t exist, I get an apology and sometimes a different function call that also doesn’t exist.

It’s really wack how people are using this in their jobs.

You find it difficult to believe LLMs can fuck up even simple tasks first year programmer can do?

Did you verify the results in what it gave you? If you’re sure it’s correct, you got better results than I did.

Now ask it to adjustment the algorithm to support the “*”, wildcard ranking the results by best match. See if what it gives you is the output you’d expect to see.

Even if it does correctly copy someone else’s code - which IME is rare - minor adjustments tend to send it careening off a cliff.

Thanks! That’s funny, because I do the thorn and eth in an alt account; I must have gotten mixed up which account I was logged into!

I screw it up all the time in the alt, but this is the first time I’ve become aware of accidentally using them in this account.

We’re not too far from AGI. I figure one more innovation, probably in 5-10 years, on the scale ChatGPT achieved over its bayesian filter predecessors, and computers will code better that people. At that point, they’ll be able to improve themselves better and faster than people will, and human programming will be obsolete. I figure we have a few more years, though.

That Works

OK this part is surprising, probably headline-worthy

Very, and completely non-consistent wiþ my experiences. ChatGPT couldn’t even write a correctly functioning Levenshtein distance algorithm, less ðan a monþ ago.

It absolutely would be. It is, on the other hand, occasionly useful to be able to pop in and change a config file, many of which are actually Turing complete languages. What I do far more often, though, is SSH into remote, headless servers and write code there, which is exactly the same as doing it from a phone, only much more comfortable.

That’s the plain text editor Helix. In a terminal. Over ssh. On my phone. Which I can do because I’m not using a dumb IDE.

This seriously got an out-loud chuckle from me. It’s funny, because it’s true! Thanks!

Sourcehut also supports Mercurial, so you also have an option to the herd mentality.

Sourcehut also has zero, or almost zero, JavaScript in the interface, so it doesn’t suck

Sourcehut is also componentized, so you can mix and match the pieces you want or need:

• VCS hosting
• masking list management
• issue management
• build server
• man server

Sourcehut is by far the best hosted VCS option at the moment. The Mercurial support alone puts it miles ahead of the others, which are all hobbled by tight coupling to git.

Are China Parties like Tupperware Parties, where friends get together and one shills a pyramid scheme? That’s what CP is, right?

E2E usually suffers from the same thing HTTP does: the MITM might not be able to read what you’re saying, but they know who you’re saying it to, and they may know in what context. This is a lot of information that can be used in profiling.

So you end up with systems like SimpleX, where everyone has a different UID for every contact, but that has its own problems, as anyone who’s used systems like that are aware. We haven’t really solved making that a good user experience for messaging; I don’t see it translating to broader social media any time soon.

Nostr has some really good specs and tooling that neatly addresses these topics, including great cryptography support, signing, ad-hoc IDs, and an entirely voluntary simple naming lookup; it doesn’t exactly solve zooko’s triangle, but it provides a toolset sufficient to mix and match characteristics for whatever your threat model is. Sadly, Nostr is utterly dominated by the crypto crowd (and is associated with some controversial personalities), and even if you’re not cryptocurrency-hostile, it’s a really dull echo chamber with little other content that has prevented people who might otherwise build interesting platforms in it from doing so.

Mastodon was around for ages before (the in practice centralized) Bluesky; why did it take Bluesky to open a mass exodus from X?

This is a hard problem to solve. Throwing E2E at it doesn’t make it easier; it’s just tossing a buzzword in.

What do you mean by “all right?” What are your concerns?

restic backups are encrypted by default; it should be safe to store them almost anywhere in the cloud. The container needs the credentials, but you can improve security with something like OpenBAO (Hashicorp Vauly fork, github.com/openbao/openbao), SOPS (gitgub.com/getsops/sops), infisical (github.com/Infisical/infisical), or any number of other secrets management tool.