Additionally, I cannot confirm the more subjective ideas - and I've been running Macbooks for almost 20 years, and specifically working with Python both for hobby, for research, professionally, for cybersecurity, etc.

I have an old 2013 laptop that is the "couch machine". It still works adequately. No issues with sleep/wake. Time machine outlasted the external HDD it was running on. I am writing this on an M1 Max, which will be 5 years old this year, and I hope I get 5 more years, it's just that good.

The python complaint I get, but it is because they ship an old python version with the OS and you have to work around that to install a different version.

Security settings can be set via the Settings app and don't require the terminal like the author stated. They can be changed via the terminal, but the golden path is just tapping a button in settings to allow the unauthorized app and typing in your password. Granted - it isn't obvious, and I only know this because over the years as notarization was added the dialogs became slowly less helpful in guiding you to the right spot, I think now in Tahoe they don't even make a mention of where you should go to allow it.

Sure, because it is from when Apple still was good.

I am writing this right now on a 2014 Mac Mini (running 10.13). Works perfect! Great machine.

The newer stuff, not so much.

I have a 2021 15"? MacBook Pro, highest end very expensive, where the battery randomly goes into rundown and drains in no time, some days it is fine. This one also can't sleep, so must do a full shutdown if I need to close the lid for more than ~10 minutes. If I close it and put it on my backpack for longer, it heats up like an oven and drains the battery. The USBC ports only work half the time at best. It's been like this since 2023, so it really only lasted two years.

I have a 2022 13" MacBook Pro where the screen is completely dead and the trackpad no longer clicks. Using it as a desktop with an external monitor and mouse, but what a disappointment. Made the mistake of buying it without Apple Care shakedown money, so can't be fixed. (Apple wanted ~$1500 to fix it, obviously not worth it).

I have a 2023 MacBook Air which is my current portable. Works ok for now, but the USBC ports are super flaky. External monitors work on a whim, sometimes, or not, which is massively annoying.

Apple quality is just pretty terrible in the 2020s. When this MacMini dies, I'm going back to a Linux desktop.

Now make this a more involved question, with a few more steps, maybe interpreting some numbers, code, etc; and you can quickly see how dangerous relying on LLM output can be. Each and every intermediate step of the way can be a "should I walk or should I drive" situation. And then the step that before that can be one too. Turtles all the way down, so to say.

I don't question that (coding) LLMs have started to be useful in my day-to-day work around the time Opus 4.5 was released. I'm a paying customer. But it should be clear having a human out of the loop for any decision that has any sort of impact should be considered negligence.

EDIT: framed the question as a riddle and all models except for Llama 4 Scout failed anyway.

The answer is right at the beginning. Current administration has the explicit goal to not have free elections going forward. It has been stated plainly, on TV. The rest is collateral damage, and an attack on critical infrastructure will be a good excuse to invade the next country, declare state of emergency or outright war and get rid of elections completely.

https://www.huffpost.com/entry/donald-trump-laura-ingraham-f...

Repeated statements by Trump and his circle claiming he’ll run in 2028. Statements by Trump that his supporters won’t ever need to vote again. That little insurrection they tried on January 6th 2021. Their current weaponization and staffing of ICE by people with questionable backgrounds and morals and deploying them against their political enemies under the pretext of illegal immigration (Texas has a bigger problem than Wisconsin For what it’s worth). Constantly praising dictatorial leaders like Putin and Xi while threatening and talking shit about Democratic allies.

So whether or not metastasizes to that point, pretending like this concern has no grounding in actual actions taken and statements uttered is wild, because this playbook isn’t new and the intended direction seems more clear than not.

- Olm/Megolm does not offer forward secrecy for group messaging

- Olm/Megolm does ensure end-to-end encryption for message data, but not for metadata.

- Federation makes it challenging to be GDPR compliant

- Synapse is very heavy, other implementations are less production ready

- For better or worse, the matrix foundation is under UK jurisdiction.

I'm sure I forget some of the nuance, but these were some of the major points. However, there are several government entities in Germany, France, Poland, etc, that can live with the limitations and DO self-host Matrix servers.

I won't go into the pair of high-severity vulns in 2025 (and the somewhat difficult mitigation) because that could hit anyone.

Megolm does provide forward secrecy - just in blocks of messages. If a message key gets stolen, an attacker could decrypt subsequent messages from that sending device until the next session begins: by default this happens either after 100 msgs have been sent, a week has elapsed, or if the room membership changes. Most folks consider this to be adequate perfect secrecy.

In terms of the Matrix Fdn being incorporated in the UK… I guess that means one shouldn’t use the Internet, given IETF is US incorporated? :)

Do Matrix clients still keep the oldest version of the Megolm ratchet they have ever received? When I last looked (around 2024), the libraries maintained by the Matrix.org core team did.

This means that, while Megolm has a ratchet that can be used to provide forward security, no Matrix implementation that I am aware of does this. This seems to me to be because other features of the Matrix specification rely on continued access to these old keys (like Megolm key backups and history sharing).

Re. security of new keys/sessions/messages after compromise of some current state (i.e. notions like post-compromise security, future secrecy):

My understanding is that, while a _sender_ will rotate Megolm sessions every 100 or so messages, recipients tend not to: clients will accept ciphertexts sent from those old sessions for an indefinite period of time. Again, I haven't been following developments in the Matrix world for a little while, so please correct me if I'm wrong.

This seems (to me) to be for similar reasons to the above: recipients keep around the recipient sessions so they can be backed up and shared with new devices (for history sharing). But (!) Matrix could get way better authentication guarantees if they just _disabled accepting messages_ from these old sessions at the same schedule as the sender stops using them.

--

These are not a unreasonable compromises (there aren't too many attempts to square this circle, and most that I'm aware of are quite academic) but it's worth making clear that just because Olm/Megolm/the Matrix spec have particular features, it doesn't mean they are used properly to give the security guarantees we would naively expect from their composition. At least, this is the case for almost all Matrix clients that I'm aware of.

It entirely depends on the client. There is nothing in the protocol which means that clients have to store old keys, but many do - mainly so they have a copy that can be backed up on the server to support migrating between devices, and for history sharing, as you say. However you absolutely could configure a locked-down Matrix client which discards megolm keys after receipt.

> My understanding is that, while a _sender_ will rotate Megolm sessions every 100 or so messages, recipients tend not to: clients will accept ciphertexts sent from those old sessions for an indefinite period of time. Again, I haven't been following developments in the Matrix world for a little while, so please correct me if I'm wrong.

Yup, this is fair - and agreed that implementations could and should discard unexpected messages in those sessions. There's nothing in the protocol that stops that (but also it's not explicitly covered in the spec).

We can fix this though; thanks for flagging it (and sorry if we missed it in the RHUL research...)

On that note, a quick correction from my side. I suggested that: "But (!) Matrix could get way better authentication guarantees if they just _disabled accepting messages_ from these old sessions at the same schedule as the sender stops using them."

But I think this is way easier said than done because (with the history sharing architecture that is currently used) it is difficult for a fresh device to meaningfully distinguish historical Megolm sessions and active ones. Other designs get around this by re-encrypting the plaintexts rather than the session keys, but this would be quite a big change.

The outputs of the IETF are RFCs. The Matrix foundation does more directly oversee the "de-facto" Matrix, so has more influence, could bow to government pressure or changing laws, etc. etc.

This ETH (i.e. Zurich) paper[0] identified several exploitable vulnerabilities (bad), which were quickly addressed by delta chat (good).

So overall, I'd see it as a good messenger, but with downsides.

[0]: https://www.usenix.org/system/files/usenixsecurity24-song-yu...

Can you elaborate? AFAIK when everything is encrypted, GDPR compliance is trivial.

> And vibe coding is fun. Even Bret Taylor, OpenAI’s chair, acknowledges it’s become a legitimate development approach.

Color me shocked! Bret, who directly profits by how his product is perceived, thinks it's legitimate???? /s

??? Do you mean biased or do you mean impartial?

My first reaction was envy. I wish human soul was mutable, too.

As long as you are still drawing breath it's never too late bud

You need cybernetics (as in the feedback loop, the habit that monitors the process of adding habits). Meditate and/or journal. Therapy is also great. There are tracking apps that may help. Some folks really like habitica/habit rpg.

You also need operant conditioning: you need a stimulus/trigger, and you need a reward. Could be as simple as letting yourself have a piece of candy.

Anything that enhances neuroplasticity helps: exercise, learning, eat/sleep right, novelty, adhd meds if that's something you need, psychedelics can help if used carefully.

I'm hardly any good at it myself but it's been some progress.

I keep gravitating to the term, "prompt adherence", because it feels like it describes the root meta-problem I have: I can set up a system, but I can't seem to get myself to follow it for more than a few days - including especially a system to set up and maintain systems. I feel that if I could crack that, set up this "habit that monitors the process of adding habits" and actually stick to it long-term, I could brute-force my way out of every other problem.

Alas.

> You know perfectly well how to achieve things without motivation.[1]

I'll also note that I'm a firm believer in removing the mental load of fake desires: If you think you want the result, but you don't actually want to do the process to get to the result, you should free yourself and stop assuming you want the result at all. Forcing that separation frees up energy and mental space for moving towards the few things you want enough.

1: https://stackingthebricks.com/how-do-you-stay-motivated-when...

Executive function.

The thing I’ve learned is for a new habit, it should have really really minimal maintenance and minimal new skill sets above the actual habit. Start with pen and paper, and make small optimizations over time. Only once you have engrained the habit of doing the thing, should you worry about optimizing it

Any notes system gets too large, and hence too complex as Is have to check a large dataset frequently. An AI could perhaps do that though..

It's exhilarating

Maybe instead of fighting it, I should build on it. Thanks!

It's a bit fascinating/unnerving to see similarities between these tools and my own context limits and that they have similar workarounds.

It's actually one of the "secret tricks" from last year, that seems to have been forgotten now that people can "afford"[0] running dozens of agents in parallel. Before everyone's focus shifted from single-agent performance to orchestration, one power move was to allow and encourage the agent to edit its own prompt/guidelines file during the agentic session, so over time and many sessions, the prompt will become tuned to both LLM's idiosyncrasies and your own expectations. This was in addition to having the agent maintain a TODO list and a "memory" file, both of which eventually became standard parts of agentic runtimes.

--

[0] - Thanks to heavy subsidizing, at least.

Only in the sense of doing circuit-bending with a sledge hammer.

> the human "soul" is a concept thats not proven yet and likely isn't real.

There are different meanings of "soul". I obviously wasn't talking about the "immortal soul" from mainstream religions, with all the associated "afterlife" game mechanics. I was talking about "sense of self", "personality", "true character" - whatever you call this stable and slowly evolving internal state a person has.

But sure, if you want to be pedantic - "SOUL.md" isn't actually the soul of an LLM agent either. It's more like the equivalent of me writing down some "rules to live by" on paper, and then trying to live by them. That's not a soul, merely a prompt - except I still envy the AI agents, because I myself have prompt adherence worse than Haiku 3 on drugs.

Remember, the Soul is just a human word, a descriptor & handle for the thing that is looking through your eyes with you. For it time doesn't exist. It is a curious observer (of both YOU and the universe outside you). Utterly neutral in most cases, open to anything and everything. It is your greatest strength, you need only say Hi to it and start a conversation with it. Be sincere and open yourself up to what is within you (the good AND the bad parts). This is just the first step. Once you have a warm welcome, the opening-up & conversation starts to flow freely and your growth will sky rocket. Soon you might discover that there are not just one of them in your but multiples, each being different natures of you. Your mind can switch between them fluently and adapt to any situation.

Maybe? So your whole premise is based on a maybe! It was a simple question, don't know where or how morality and behavior comes into play..

> "The human brain is mutable, the human "soul" is a concept thats not proven yet and likely isn't real."

The soul is "a concept that's not proven yet." It's unproven because there's no convincing evidence for the proposition. By definition, in the absence of convincing evidence, the null hypothesis of any proposition is presumed to be more likely. The presumed likelihood of the null hypothesis is not a positive assertion which creates a burden of proof. It's the presumed default state of all possible propositions - even those yet to be imagined.

In other words, pointing out 'absence of evidence' is not asserting 'evidence of absence'. See: Russell's Teapot and Sagan's Dragon (https://en.wikipedia.org/wiki/Russell%27s_teapot)

Whereas with Python, even in the latest version, you're already looking at atleast 10x the amount of startup latency in practice.

Note: This is excluding the actual time that is made for the network call, which can of course also add quiete some milliseconds, depending on how far on planet earth your destination is.

Compare:

    import requests
    print(requests.get("http://localhost:3000").text)

    package main

    import (
      "fmt"
      "io"
      "net/http"
     )

    func main() {
        resp, _ := http.Get("http://localhost:3000")
        defer resp.Body.Close()
        body, _ := io.ReadAll(resp.Body)
        fmt.Println(string(body))
    }

    python3:  0.08s user 0.02s system 91% cpu 0.113 total
    go 0.00s user 0.01s system 72% cpu 0.015 total

I wrote another that counts the lines in a file, and tested it against https://www.gutenberg.org/cache/epub/2600/pg2600.txt

I get:

    python 0.03s user 0.01s system 83% cpu 0.059 total
    go 0.00s user 0.00s system 80% cpu 0.010 total

I believe in the past people have looked at putting the standard library in a zip file instead of splatted out into a bunch of files in a dirtree. In that case, I think python would just do a few stats, find the zipfile, loaded the whole thing into RAM, and then index into the file.

"If python was implemented totally different it might be fast" - sure, but it's not!

It's tooling agnostic and there are a couple ways to generate them, but the easiest it to just use pants build.

Pants also does dependency traversal (that's the main reason we started using it, deploying a microservices monorepo) so it only packages the necessary modules.

I haven't profiled it yet for cold starts, maybe I'll test that real quick.

https://www.pantsbuild.org/dev/docs/python/overview/pex

Edit: just ran it on a hello world with py3.14 on m3 macbook pro, about 100 +/-30 ms for `python -m hello` and 300-400 (but wild variance) for executing the pex with `./hello/binary.pex`.

I'm not sure if a pants expert could eke out more speed gains and I'm also not sure if this strategy would win out with a lot of dependencies. I'm guessing the time required to stat every imported file pales in comparison to the actual load time, and with pex, everything needs to be unzipped first.

Pex is honestly best when you want to build and distribute an application as a single file (there are flags to bundle the python interpreter too).

The other option is mypyc, though again that seems to mostly speed up runtime https://github.com/mypyc/mypyc

Now if I use `python -S` (disables `import site` on initialization), that gets down to ~15ms execution time for hello world. But that gain gets killed as soon as you start trying to import certain modules (there is a very limited set of modules you can work with and still keep speedup. So if you whole script is pure python with no imports, you could probably have a 20ms cold start).